Deep slate report 20260922

Which reports actually fed the diagnosis. A source shown as missing was not silently assumed.

High-level causal issue register

Source	Status	Evidence supplied	Path	Bytes
prediction review	loaded	official outcomes and scoring errors	/app/output/reports/prediction_review_20260922.json	401690
forensic deep audit	loaded	probability, provenance, starter and decision contracts	/app/output/reports/PREDICTION_SYSTEM_DEEP_AUDIT_20260922.json	255950
continuous learning	loaded	pipeline-stage quality, calibration, probability lineage, score distribution, and official starter comparisons	/app/output/learning/continuous_learning_daily_20260922.json	2542724
detailed report	loaded	full game and simulation rendering	/app/output/reports/detailed_20260922.html	379842
ultra detailed player audit	loaded	player-level prediction trace	/app/output/reports/20260922_ultra_detailed_player_audit.html	34855359
structural audit	loaded	all invariant findings	/app/output/reports/audit_report_20260922.html	2002677
integrity revision audit	missing	post-Luna remediation proof for immutable revision artifacts	/app/output/reports/audit_report_20260922_integrity_revision.html	n/a

Deterministic findings across data integrity, model validation, calibration, score distribution, pipeline stages, simulation, and coverage. Critical is reserved for broken evidence or an unexplained contract violation; high marks material model/process risk. Each row pairs the action to take (Do this) with the wrong conclusion to avoid (Don't conclude).

Cross-slate trends (persistent-pattern memory)

Escalated persistent patterns

A finding that has repeated at or beyond the escalation streak is no longer single-slate noise. It carries its target registered lever and an action-oriented guardrail.

Finding streaks across the window

Metric direction across the window

Pipeline stage regression

Every stored stage is scored against the same official finals. A later recovery remains visible; it does not erase an unstable intermediate flip.

Probability lineage and calibration

All columns are pick-side probabilities, so away favorites are not accidentally compared using home-side semantics.

Score distribution and margin compression

Narrow predicted margins next to wide actual margins is the signature the score-variance tail is too tight.

Per-game evidence coverage

Every game, in depth (15)

Severity	Domain	Finding and evidence	Affected	Action & guardrail
CRITICAL	probability_lineage	Published probability differs materially without a recorded transformation 1 game(s) have an unexplained ≥2pp calibrated-to-published delta	WAS@DET	Do thisBlock any promotion until the missing probability transformation is recorded and reconciled. Don't concludeDo not assume the probability governor was bypassed until you have traced the pick-side math.
HIGH	calibration	Published confidence buckets were poorly aligned on this slate weighted absolute calibration error 0.1881 across 15 games	ARI@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@SEA, LAA@OAK, MIA@CHC, MIL@PHI, MIN@SF, NYM@TEX, SD@LAD, STL@PIT, TB@NYY, TB@NYY, WAS@DET	Do thisWatch the same confidence buckets over the full rolling window before changing any calibration weight. Don't concludeA single slate has far too few games to establish a stable calibration curve.
HIGH	model_validation	Validated run models explain very little holdout variance home/away/total R² = 0.0172/0.0294/0.0090	ARI@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@SEA, LAA@OAK, MIA@CHC, MIL@PHI, MIN@SF, NYM@TEX, SD@LAD, STL@PIT, TB@NYY, TB@NYY, WAS@DET	Do thisTreat exact run scores as low-confidence estimates, and look for stronger features or a better target before tuning the score tails. Don't concludeA small positive R² only means the model beats guessing the average; it does not make the exact scores reliable.
MEDIUM	calibration	Most high-raw picks lost before/after shrinkage 0/3 games with raw pick probability ≥0.65 won	CIN@ATL, HOU@SEA, MIA@CHC	Do thisTrack how often high-raw-probability picks lose across multiple slates, and keep the raw to calibrated to published lineage. Don't concludeThis one slate is not enough to justify a new per-bucket shrinkage rule.
INFO	coverage	Every settled game received structural QA 16/15 games structurally audited	ARI@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@SEA, LAA@OAK, MIA@CHC, MIL@PHI, MIN@SF, NYM@TEX, SD@LAD, STL@PIT, TB@NYY, TB@NYY, WAS@DET	Do thisKeep the per-game audit receipt in every report. Don't concludeFull audit coverage does not cancel the findings raised inside the audited games.
INFO	simulation	Simulation produced no winner-direction intervention on this slate flip candidates 0; published flips 0	ARI@COL, CHW@KC, CIN@ATL, CLE@BOS, HOU@SEA, LAA@OAK, MIA@CHC, MIL@PHI, MIN@SF, NYM@TEX, SD@LAD, STL@PIT, TB@NYY, TB@NYY, WAS@DET	Do thisMeasure the simulation's contribution across rolling slates, separating score realization from winner flipping. Don't concludeZero winner flips is not a defect and does not mean the simulation did no score-reconciliation work.

Finding	Persistence	Target lever(s)	Direction	Guardrail
simulation_inert_on_slate	11 consecutive · 11 in window	simulation.allow_score	confirm the simulation module is running, not silently no-oping	simulation_inert_on_slate has persisted 11 consecutive slates; this is no longer single-slate noise. Treat it as an actionable pattern and raise the code/diagnostic action named; do not force a parameter move.
single_slate_calibration_instability	4 consecutive · 8 in window	confidence.empirical_max_weight, confidence.base_cap, confidence.moderate_cap	increase empirical shrinkage	single_slate_calibration_instability has persisted 4 consecutive slates; this is no longer single-slate noise. Treat it as an actionable pattern and propose a bounded shadow experiment against a target lever.
weak_run_model_signal	11 consecutive · 11 in window	training.xgb.n_estimators, training.xgb.max_depth, training.xgb.learning_rate	investigate features/target via isolated training backtest (manual)	weak_run_model_signal has persisted 11 consecutive slates; this is no longer single-slate noise. Treat it as an actionable pattern and raise the code/diagnostic action named; do not force a parameter move.

Finding	Consecutive slates	Incidence	Present now	Fix type
high_raw_probability_cluster	1	3/11	yes	param
pipeline_winner_instability	0	5/11	no	code/diagnostic
score_margin_compression	0	1/11	no	param
simulation_inert_on_slate	11	11/11	yes	code/diagnostic
single_slate_calibration_instability	4	8/11	yes	param
weak_run_model_signal	11	11/11	yes	code/diagnostic

Metric	Direction	Series (old → new)
winner_accuracy	regressing	[0.5333, 0.6, 0.7, 0.4, 0.6667, 0.4444, 0.6667, 0.7333, 0.6, 0.3333, 0.4]
weighted_calibration_error	improving	[0.1711, 0.0827, 0.2812, 0.2795, 0.1661, 0.1016, 0.1247, 0.3458, 0.3184, 0.2028, 0.1881]
margin_bias	regressing	[-2.667, -1.8, -0.4, -1.667, -1.133, -2.111, -2.0, -0.667, -1.533, -2.667, -1.867]
team_run_error_on_losses	regressing	[6.143, 5.0, 6.333, 5.556, 6.4, 3.6, 5.2, 6.5, 6.5, 3.0, 6.222]

Stage	Winner hits	Accuracy	Team-run MAE	Total-run MAE	Margin MAE
historical_ml	7/15	46.7%	2.907	4.061	3.428
variance_applied	6/15	40%	2.967	4.2	3.533
enhanced_prediction	7/15	46.7%	3.2	4.133	4
embedded_savant_run_environment	6/15	40%	3.167	4.2	4.067
fundamental_edge_target_review	6/15	40%	3.2	4.267	4.133
capped_target	6/15	40%	3.2	4.267	4.133
simulation_result	6/15	40%	3.2	4.267	4.133
published_final	6/15	40%	3.2	4.267	4.133

Transition	Δ accuracy	Δ hits	Δ team-run MAE	Δ margin MAE	Games whose winner/tie changed
historical_ml → variance_applied	-0.0667	-1	0.0601	0.1056	LAA@OAK: OAK → TIE; NYM@TEX: TEX → TIE
variance_applied → enhanced_prediction	0.0667	1	0.2333	0.4667	CHW@KC: KC → CHW; CLE@BOS: BOS → CLE; LAA@OAK: TIE → LAA; MIL@PHI: PHI → MIL; NYM@TEX: TIE → TEX; TOR@BAL: TOR → BAL
enhanced_prediction → embedded_savant_run_environment	-0.0667	-1	-0.0333	0.0667	CLE@BOS: CLE → TIE
embedded_savant_run_environment → fundamental_edge_target_review	0	0	0.0333	0.0666	CLE@BOS: TIE → BOS
fundamental_edge_target_review → capped_target	0	0	0	0
capped_target → simulation_result	0	0	0	0
simulation_result → published_final	0	0	0	0

Game	Outcome	Raw model	Calibrated	Governor	Score coherence	Published	Explanation
ARI@COL	hit	0.6408	0.6224	0.6224	0.5982	0.56	score coherence followed by matchup-recency guardrail
CHW@KC	hit	0.5	0.5073	0.5073	0.5073	0.5053	published probability matches the calibrated/governed value
CIN@ATL	miss	0.6937	0.699	0.699	0.699	0.6	score coherence followed by matchup-recency guardrail
CLE@BOS	miss	0.5	0.5073	0.5073	0.5073	0.5053	published probability matches the calibrated/governed value
HOU@SEA	miss	0.6546	0.6409	0.6409	0.6409	0.6	score coherence followed by matchup-recency guardrail
LAA@OAK	miss	0.5132	0.5602	0.5602	0.5602	0.53	score coherence followed by matchup-recency guardrail
MIA@CHC	miss	0.6927	0.6546	0.6546	0.6501	0.53	score coherence followed by matchup-recency guardrail
MIL@PHI	miss	0.537	0.6768	0.6768	0.6768	0.6	score coherence followed by matchup-recency guardrail
MIN@SF	hit	0.5367	0.5232	0.5232	0.5232	0.5154	score coherence followed by matchup-recency guardrail
NYM@TEX	miss	0.5722	0.6009	0.6009	0.6009	0.6009	published probability matches the calibrated/governed value
SD@LAD	hit	0.6139	0.5641	0.5641	0.5641	0.53	score coherence followed by matchup-recency guardrail
STL@PIT	hit	0.5625	0.5535	0.5535	0.5535	0.5387	score coherence followed by matchup-recency guardrail
TB@NYY	hit	0.5842	0.5483	0.5483	0.5483	0.5349	score coherence followed by matchup-recency guardrail
TB@NYY	miss	0.5501	0.5286	0.5286	0.5286	0.5206	score coherence followed by matchup-recency guardrail
WAS@DET	miss	0.6101	0.5929	0.5929	0.5835	0.5709	no recorded transformation explains the publication delta

Sides analyzed	Broke on pitching	Broke on offense	Both	Variance	Mean starter ERA error	Mean lineup runs vs proj
30	3	11	3	13	-0.32	-0.97

Game	Trained p(home)	Blended p(home)	Blend favoured	Published p(home)	Published side	Artifact says winner changed	Outcome
CHW@KC	0.616	0.554	home	0.495	away	yes	hit

Slate	Games	Winner acc	Mean starter ERA error	Mean lineup runs vs proj	Sides broke on pitching	Sides broke on offense
20260921	3	0.333	-1.08	-0.50
20260920	15	0.600	-0.89	-1.00	2	10
20260919	15	0.733	1.36	-0.10	8	7
20260918	15	0.667	3.95	0.30	6	5
20260917	9	0.444	2.16	-0.22	2	3
20260916	15	0.667	2.09	-0.23	4	5
20260915	15	0.400	-1.58	-0.70	1	12
20260914	10	0.700	-0.48	-0.20	2	6
20260913	15	0.600	-0.07	0.03	2	8
20260912	15	0.533	1.35	0.57	6	7
20260911	15	0.667	0.99	0.00	4	9
20260910	5	0.600	-3.19	-1.70		8

Game	Pred margin	Actual margin	Pred − actual
ARI@COL	1	5	-4
CHW@KC	1	3	-2
CIN@ATL	3	4	-1
CLE@BOS	1	1	0
HOU@SEA	2	7	-5
LAA@OAK	1	2	-1
MIA@CHC	2	6	-4
MIL@PHI	3	2	1
MIN@SF	2	1	1
NYM@TEX	2	3	-1
SD@LAD	2	7	-5
STL@PIT	1	2	-1
TB@NYY	1	2	-1
TB@NYY	1	5	-4
WAS@DET	1	2	-1

ARI@COLHIT

Predicted: ARI 6-5 (conf -0.12, moderate)Actual: ARI 7-2Brier 0.194total bias +2

Root cause: Solid correct call

Evidence-backed diagnosis

published system edge 0.12 (moderate); this is not the half-scale distance from 50%
team-run absolute error 4; total bias +2
pregame home-minus-away evidence: season -1.052, recent -0.534, pitcher -0.043, composite -0.987
confidence pipeline reached 0.622 before later publication/recency controls; published home probability was 0.440
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

COL offense over-projected by 3

Team	Starter	Projected	Actual	ER err
ARI	Michael Soroka	2 ER / 15 outs	1 ER / 21 outs	-1
COL	Kyle Freeland	5 ER / 14 outs	5 ER / 9 outs	0

Decision internals: run factors home×0.983 / away×1.026; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Nolan Arenado (ARI 3B): 3-for-5, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 2.18]
Hunter Goodman (COL C): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Ketel Marte (ARI DH): 1-for-5, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]
Mickey Moniak (COL DH): 1-for-3, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.93]
Gabriel Moreno (ARI C): 1-for-4, 0 HR, 2 RBI (projected 2.0 H, 0.0 HR) [run impact -0.8]

Starter miss: Kyle Freeland (COL) posted 15.00 ERA vs projected 9.64 (5 ER in 9 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

What to do: Variance — no model change is warranted for this start.

Pitcher versus batter: what actually broke this game

Per-pitch matchup fields are not populated in the current feeds, so this is reconstructed from the settled box: each starter's workload-adjusted ERA, the bullpen share behind him, and the opposing lineup's real production against its projection, with the hitters who beat that projection named.

AWAY pitching

No settled pitching line for this club. The forecast named Michael Soroka (ARI) at a projected 3.6 ERA, but the per-player feed carries no pitching row for him in this game, so the actual-versus-projected comparison cannot be made on this side. That is a source-coverage gap in the settled box, not a forecasting failure.

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 2 runs on 3 hits and 1 HR, against a projected 4 runs / 9 hits (-2 runs vs projection).

Who did the damage to him:

Mickey Moniak: 1-for-3, 1 HR, 1 RBI [+0.93 runs vs projection] season xwOBA 0.277, weak - a good day, not a model error
Troy Johnston: 1-for-2, 0 HR, 0 RBI [+0.31 runs vs projection] season xwOBA 0.323, above average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Kyle Freeland (COL) threw 9 outs, 5 ER = 15 ERA against a projected 9.64 ERA (5.36 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The AWAY lineup he faced produced 5 runs on 11 hits and 1 HR, against a projected 4 runs / 8 hits (1 runs vs projection).

Who did the damage to him:

Nolan Arenado: 3-for-5, 1 HR, 1 RBI [+2.18 runs vs projection] season xwOBA 0.293, below average - a good day, not a model error
Tim Tawa: 1-for-2, 0 HR, 2 RBI [+0.64 runs vs projection] season xwOBA 0.306, average - a good day, not a model error
Geraldo Perdomo: 3-for-5, 0 HR, 1 RBI [+0.61 runs vs projection] season xwOBA 0.327, above average - real projection error

Verdict: both: the starter was hit hard AND the lineup underproduced

Starting pitchers & relievers: actual / projected + context (ERA is workload-aware: ER over 9 innings, so a short blow-up is not read like a quality start; ERA error in parentheses when material. exp BB and exp SO are the continuous rate expectations. The integer beside each is a simulation draw reconciled to the opposing simulated batting box, so it is not the line that ranks pitchers: on the 2026-09-07 slate the integers published 39 walks and 121 strikeouts against 51 and 130 real, while these expectations landed on 50.4 and 132.6.)

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
COL	Kyle Freeland (Starter)	9 / 14	15 / 9.64 (+5.36)	5 / 5	9 / 6	0 / 2	1 / 2	1.34	2 / 2	4.23	venue home: 5.82 vs 6.96 ERA (ok) · opp run factor 1.03 (average) · 24 prior appearances this season

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Nolan Arenado (3B)	ARI	5 / 4	3 / 1	1 / 0	0 / 0	1 / 0	1 / 1	2 / 1	0 / 0	1 / 1	2.18	0.293 below average	good day
Hunter Goodman (C)	COL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-1.4	0.32 average	no signal
Ketel Marte (DH)	ARI	5 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	0 / 0	-0.93	0.345 elite	bad day
Mickey Moniak (DH)	COL	3 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 0	0 / 0	2 / 1	0.93	0.277 weak	good day
Gabriel Moreno (C)	ARI	4 / 4	1 / 2	1 / 1	0 / 0	0 / 0	2 / 1	1 / 1	0 / 1	0 / 1	-0.8	0.363 elite	bad day
Kyle Karros (3B)	COL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	-0.8	0.341 elite	bad day
TJ Rumfield (1B)	COL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	1 / 0	-0.8	0.336 above average	bad day
Jake McCarthy (CF)	COL	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	2 / 1	-0.78	0.325 above average	bad day
Tim Tawa (LF)	ARI	2 / 3	1 / 1	1 / 0	0 / 0	0 / 0	2 / 0	0 / 0	1 / 0	0 / 1	0.64	0.306 average	no signal
Geraldo Perdomo (SS)	ARI	5 / 4	3 / 1	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	0 / 1	0 / 0	0.61	0.327 above average	projection too low
Connor Norby (2B)	COL	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	1 / 1	-0.47	0.306 average	—
Zac Veen (LF)	COL	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.286 below average	—
Jordan Lawlar (CF)	ARI	4 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	0.31	0.314 average	—
Corbin Carroll (RF)	ARI	4 / 4	1 / 1	1 / 0	0 / 1	0 / 0	1 / 1	0 / 1	0 / 0	0 / 1	-0.31	0.343 elite	—
Cole Carrigg (SS)	COL	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	-0.31	0.311 average	—
Troy Johnston (RF)	COL	2 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0.31	0.323 above average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-1.052	positive favors home; negative favors away
Recent form	-0.534	positive favors home; negative favors away
Starting-pitcher matchup	-0.043	positive favors home; negative favors away
Composite	-0.987	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	5.79-4.15	n/a	Base ML historical model prediction
variance_applied	6-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	6-5	0.56	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	6-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-5	n/a	Capped at max 12 realistic runs
simulation_result	6-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	5.79-4.15	6-4	0.358	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	6-4	6-5	-1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	6-5	6-5	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-5	6-5	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-5	6-5	-0	Capped at max 12 realistic runs
simulation_result	6-5	6-5	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
COL	Troy Johnston	batter / RF	worse	predicted_RBI 2.83
ARI	Tim Tawa	batter / LF	better	predicted_RBI 2.568
ARI	Geraldo Perdomo	batter / SS	worse	predicted_H 1.85
COL	Zac Veen	batter / LF	deboosted	predicted_AB 1.5
COL	Kyle Freeland	pitcher / Starter	worse	predicted_IP -1.467
ARI	Gabriel Moreno	batter / C	worse	predicted_H -1.46
ARI	Nolan Arenado	batter / 3B	worse	predicted_H -1.27
COL	Hunter Goodman	batter / C	worse	predicted_H -1.17

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	ARI pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	ARI
info	CURRENT_PITCHER_DUPLICATION	COL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	COL
info	CURRENT_FINALIZED_MATCH_BOX	ARI model-prediction batting totals differ from final simulation box score (H: current=11, sim=7; 2B: current=2, sim=0; 3B: current=1, sim=0; HR: current=1, sim=3; RBI: current=5, sim=6; R: current=5, sim=6; SO: current=6, sim=8; SF: current=0, sim=1).	ARI
info	CURRENT_PLAYER_MATCH_BOX	ARI current finalized hitter lines are model predictions and differ from the simulation box score: Ketel Marte: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, SF current=0 box=1; Ildemaro Vargas: H current=2 box=1, 2B current=1 box=0, RBI current=1 box=0, SO current=0 box=1; Jordan Lawlar: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=1, R current=0 box=1; Corbin Carroll: H current=1 box=0, 3B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Tim Tawa: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=3, R current=0 box=1; Jose Fernandez: HR current=0 box=1, RBI current=0 box=1, R current=0 box=1, BB current=0 box=1; Gabriel Moreno: H current=2 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Nolan Arenado: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, R current=1 box=0	ARI
info	CURRENT_FINALIZED_MATCH_BOX	COL model-prediction batting totals differ from final simulation box score (AB: current=34, sim=32; H: current=9, sim=5; 2B: current=2, sim=0; 3B: current=0, sim=1; R: current=4, sim=5; BB: current=3, sim=4; SO: current=8, sim=6; HBP: current=0, sim=1).	COL
info	CURRENT_PLAYER_MATCH_BOX	COL current finalized hitter lines are model predictions and differ from the simulation box score: Jake McCarthy: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Cole Carrigg: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Kyle Karros: AB current=4 box=3, 3B current=0 box=1, SO current=1 box=0; Mickey Moniak: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Hunter Goodman: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Connor Norby: RBI current=0 box=1, R current=0 box=1, BB current=1 box=0, SO current=1 box=2; TJ Rumfield: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, R current=0 box=1; Troy Johnston: H current=1 box=2, HR current=0 box=1, RBI current=0 box=3, R current=0 box=1	COL
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

CHW@KCHIT

Predicted: CHW 5-4 (conf -0.01, toss_up)Actual: CHW 14-11Brier 0.245total bias -16

Root cause: Toss-up that went our way

Evidence-backed diagnosis

published system edge 0.01 (toss_up); this is not the half-scale distance from 50%
team-run absolute error 16; total bias -16
pregame home-minus-away evidence: season -0.833, recent -0.642, pitcher +0.025, composite -0.744
confidence pipeline reached 0.507 before later publication/recency controls; published home probability was 0.495
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs under-projected by 16 (9 vs 25)
CHW offense under-projected by 9
KC offense under-projected by 7
Daniel Lynch IV (KC) allowed 8 more ER than projected (9 vs 1)

Team	Starter	Projected	Actual	ER err
CHW	Anthony Kay	3 ER / 15 outs	4 ER / 12 outs	1
KC	Daniel Lynch IV	1 ER / 17 outs	9 ER / 12 outs	8

Decision internals: run factors home×0.968 / away×0.943; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Chase Meidroth (CHW 2B): 3-for-4, 1 HR, 5 RBI (projected 1.0 H, 0.0 HR) [run impact 2.2]
Vinnie Pasquantino (KC 1B): 3-for-5, 1 HR, 5 RBI (projected 1.0 H, 0.0 HR) [run impact 2.18]
Carter Jensen (KC C): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.26]
Jac Caglianone (KC RF): 2-for-5, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 0.78]
Bobby Witt Jr. (KC SS): 3-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 0.61]

Starter miss: Daniel Lynch IV (KC) posted 20.25 ERA vs projected 1.59 (9 ER in 12 outs); he is materially worse pitching home (5.85 ERA there vs 4.22 on the other side), a venue-role weakness the model under-weighted.

What to do: Enable/raise the venue-split weight (context.venue_adjustment_weight) and validate in shadow; the split is credible for this starter.

Pitcher versus batter: what actually broke this game

AWAY pitching

No settled pitching line for this club. The forecast named Anthony Kay (CHW) at a projected 5.4 ERA, but the per-player feed carries no pitching row for him in this game, so the actual-versus-projected comparison cannot be made on this side. That is a source-coverage gap in the settled box, not a forecasting failure.

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 10 runs on 12 hits and 2 HR, against a projected 4 runs / 9 hits (6 runs vs projection).

Who did the damage to him:

Vinnie Pasquantino: 3-for-5, 1 HR, 5 RBI [+2.18 runs vs projection] season xwOBA 0.333, above average - real projection error
Jac Caglianone: 2-for-5, 0 HR, 0 RBI [+0.78 runs vs projection] season xwOBA 0.357, elite - real projection error
Bobby Witt Jr.: 3-for-4, 0 HR, 0 RBI [+0.61 runs vs projection] season xwOBA 0.376, elite - real projection error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Daniel Lynch IV (KC) threw 12 outs, 9 ER = 20.25 ERA against a projected 1.59 ERA (18.66 ERA error)

Bullpen behind him: 0 ER over 3 outs.

The AWAY lineup he faced produced 6 runs on 6 hits and 2 HR, against a projected 5 runs / 4 hits (1 runs vs projection).

Who did the damage to him:

Chase Meidroth: 3-for-4, 1 HR, 5 RBI [+2.2 runs vs projection] season xwOBA 0.296, below average - a good day, not a model error

Verdict: pitching: the starter gave up runs at a far worse rate than projected

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
KC	Daniel Lynch IV (Starter)	12 / 17	20.25 / 1.59 (+18.66)	9 / 1	7 / 3	3 / 0	3 / 3	1.32	2 / 4	2.03	venue home: 5.85 vs 4.22 ERA (worse here) · opp run factor 0.94 (average) · 54 prior appearances this season
KC	Alex Lange (Middle Reliever (High Leverage))	3 / 4	0 / 20.25 (-20.25)	0 / 3	0 / 4	0 / 0	0 / 1	0.5	2 / 0	0.82

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Chase Meidroth (2B)	CHW	4 / 4	3 / 1	0 / 0	0 / 0	1 / 0	5 / 0	2 / 1	1 / 0	0 / 1	2.2	0.296 below average	good day
Vinnie Pasquantino (1B)	KC	5 / 4	3 / 1	1 / 0	0 / 0	1 / 0	5 / 1	3 / 1	0 / 0	0 / 0	2.18	0.333 above average	projection too low
Carter Jensen (C)	KC	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	1 / 1	-1.26	0.31 average	no signal
Jac Caglianone (RF)	KC	5 / 4	2 / 1	1 / 0	0 / 0	0 / 0	0 / 1	2 / 1	0 / 0	1 / 1	0.78	0.357 elite	projection too low
Bobby Witt Jr. (SS)	KC	4 / 4	3 / 1	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0.61	0.376 elite	projection too low
Nick Loftin (LF)	KC	4 / 4	1 / 1	0 / 0	0 / 0	1 / 0	2 / 0	1 / 0	0 / 1	0 / 0	0.6	0.293 below average	good day
Kyle Isbel (CF)	KC	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	3 / 1	-0.47	0.255 weak	—
Miguel Vargas (3B)	CHW	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0 / 1	-0.31	0.381 elite	—
Michael Massey (2B)	KC	5 / 4	1 / 1	1 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0 / 1	0.31	0.307 average	—
Maikel Garcia (3B)	KC	2 / 4	0 / 1	0 / 1	0 / 0	0 / 0	1 / 0	3 / 0	3 / 0	0 / 0	0.21	0.31 average	—
Kyle Teel (C)	CHW	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 0	2 / 1	-0.14	0.308 average	—
Braden Montgomery (RF)	CHW	4 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	1 / 0	0.14	0.325 above average	—
Munetaka Murakami (1B)	CHW	3 / 4	1 / 1	0 / 0	0 / 0	1 / 1	2 / 1	2 / 1	1 / 1	1 / 1	0	0.34 elite	—
Colson Montgomery (SS)	CHW	5 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	3 / 1	0	0.303 average	—
Salvador Perez (DH)	KC	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	1 / 1	0	0.31 average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.833	positive favors home; negative favors away
Recent form	-0.642	positive favors home; negative favors away
Starting-pitcher matchup	0.025	positive favors home; negative favors away
Composite	-0.744	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.09-4.92	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	0.505	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	n/a	Capped at max 12 realistic runs
simulation_result	5-4	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.09-4.92	4-5	-0.174	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	5-4	2	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	5-4	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	5-4	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	5-4	-0	Capped at max 12 realistic runs
simulation_result	5-4	5-4	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
CHW	Randal Grichuk	batter / LF	better	predicted_AB 2.5
CHW	Munetaka Murakami	batter / 1B	worse	predicted_SO -1.558
CHW	Tommy Pham	batter / DH	better	predicted_AB 1.5
KC	Vinnie Pasquantino	batter / 1B	worse	predicted_H 1.5
CHW	Colson Montgomery	batter / SS	better	predicted_SO -1.469
KC	Bobby Witt Jr.	batter / SS	worse	predicted_H -1.29
CHW	Brenton Doyle	batter / CF	worse	predicted_AB 1.08
KC	Salvador Perez	batter / DH	worse	predicted_H 1.03

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	CHW pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	CHW
info	CURRENT_PITCHER_DUPLICATION	KC pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	KC
info	CURRENT_FINALIZED_MATCH_BOX	CHW model-prediction batting totals differ from final simulation box score (AB: current=31, sim=35; H: current=7, sim=9; RBI: current=4, sim=5; BB: current=3, sim=4; SO: current=8, sim=5; SF: current=0, sim=1).	CHW
info	CURRENT_PLAYER_MATCH_BOX	CHW current finalized hitter lines are model predictions and differ from the simulation box score: Tommy Pham: AB current=3 box=4, 2B current=0 box=1, R current=0 box=1, BB current=0 box=1; Randal Grichuk: AB current=4 box=5, H current=1 box=2, RBI current=1 box=2, R current=0 box=1; Munetaka Murakami: H current=1 box=0, HR current=1 box=0, BB current=1 box=0, SO current=1 box=0; Chase Meidroth: H current=1 box=0, R current=1 box=0; Kyle Teel: AB current=3 box=4, RBI current=1 box=0, R current=1 box=0, SO current=1 box=2; Braden Montgomery: H current=0 box=1; Brenton Doyle: AB current=3 box=4, H current=1 box=0; Miguel Vargas: H current=1 box=2, 2B current=1 box=0	CHW
info	CURRENT_FINALIZED_MATCH_BOX	KC model-prediction batting totals differ from final simulation box score (AB: current=35, sim=37; H: current=9, sim=10; 2B: current=2, sim=1; HR: current=1, sim=0; BB: current=3, sim=0; SO: current=6, sim=5).	KC
info	CURRENT_PLAYER_MATCH_BOX	KC current finalized hitter lines are model predictions and differ from the simulation box score: Bobby Witt Jr.: H current=1 box=0, 2B current=1 box=0, R current=1 box=0, BB current=1 box=0; Carter Jensen: AB current=4 box=5, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Maikel Garcia: 2B current=1 box=0, RBI current=0 box=1, R current=0 box=1; Jac Caglianone: H current=1 box=2, 2B current=0 box=1, RBI current=1 box=0, R current=1 box=0; Nick Loftin: H current=1 box=0, BB current=1 box=0, SO current=0 box=1; Vinnie Pasquantino: H current=1 box=2, SO current=0 box=1; Salvador Perez: H current=1 box=2, R current=0 box=1; Kyle Isbel: AB current=3 box=4, RBI current=0 box=1, R current=0 box=1, SO current=1 box=0	KC
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

CIN@ATLMISS

Predicted: ATL 6-9 (conf +0.20, strong)Actual: CIN 4-0Brier 0.36total bias +11

Root cause: Starting-pitcher projection was wrong

Evidence-backed diagnosis

published system edge 0.20 (strong); this is not the half-scale distance from 50%
team-run absolute error 11; total bias +11
pregame home-minus-away evidence: season +1.699, recent +1.245, pitcher +0.059, composite +1.917
confidence pipeline reached 0.699 before later publication/recency controls; published home probability was 0.600
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 11 (15 vs 4)
ATL offense over-projected by 9
Brandon Williamson (CIN) allowed 5 fewer ER than projected (0 vs 5)
committed to ATL at 0.20 confidence and it lost

Team	Starter	Projected	Actual	ER err
CIN	Brandon Williamson	5 ER / 12 outs	0 ER / 18 outs	-5
ATL	JR Ritchie	2 ER / 14 outs	1 ER / 15 outs	-1

Decision internals: run factors home×1.042 / away×0.982; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Elly De La Cruz (CIN SS): 2-for-3, 1 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.42]
Matt Olson (ATL 1B): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.26]
Ronald Acuña Jr. (ATL RF): 0-for-1, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Eugenio Suárez (CIN 3B): 1-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.93]
Michael Harris II (ATL CF): 2-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 0.78]

Starter miss: Brandon Williamson (CIN) posted 0.00 ERA vs projected 11.25 (0 ER in 18 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

What to do: Variance — no model change is warranted for this start.

Pitcher versus batter: what actually broke this game

AWAY pitching

Brandon Williamson (CIN) threw 18 outs, 0 ER = 0 ERA against a projected 11.25 ERA (-11.25 ERA error) only 10 prior appearances in the covered season - the projection had little of his own record to stand on

Bullpen behind him: 0 ER over 9 outs.

The HOME lineup he faced produced 0 runs on 7 hits and 0 HR, against a projected 5 runs / 8 hits (-5 runs vs projection).

Who did the damage to him:

Michael Harris II: 2-for-4, 0 HR, 0 RBI [+0.78 runs vs projection] season xwOBA 0.339, above average - real projection error
Drake Baldwin: 2-for-3, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.367, elite - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

JR Ritchie (ATL) threw 15 outs, 1 ER = 1.8 ERA against a projected 3.86 ERA (-2.06 ERA error)

Bullpen behind him: 0 ER over 3 outs.

The AWAY lineup he faced produced 4 runs on 10 hits and 3 HR, against a projected 4 runs / 8 hits (0 runs vs projection).

Who did the damage to him:

Elly De La Cruz: 2-for-3, 1 HR, 2 RBI [+1.42 runs vs projection] season xwOBA 0.375, elite - real projection error
Eugenio Suárez: 1-for-4, 1 HR, 1 RBI [+0.93 runs vs projection] season xwOBA 0.298, below average - a good day, not a model error
Dane Myers: 1-for-4, 1 HR, 1 RBI [+0.6 runs vs projection] season xwOBA 0.31, average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
CIN	Brandon Williamson (Starter)	18 / 12	0 / 11.25 (-11.25)	0 / 5	3 / 6	0 / 2	0 / 1	4.14	2 / 1	3.97	venue away: 3.31 vs 8.47 ERA (ok) · opp run factor 1.04 (strong) · 10 prior appearances this season
CIN	Brock Burke (Middle Reliever (High Leverage))	3 / 3	0 / 27 (-27.00)	0 / 3	1 / 3	0 / 1	1 / 1	0.49	1 / 2	1
CIN	Emilio Pagán (Closer)	3 / 3	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0.3	1 / 0	1.27
CIN	Tejay Antone (Middle Reliever)	3 / 1	0 / 0	0 / 0	2 / 0	0 / 0	1 / 1	0.35	1 / 1	1.14
ATL	JR Ritchie (Starter)	15 / 14	1.8 / 3.86	1 / 2	4 / 2	1 / 0	3 / 2	2.48	8 / 7	4.72	venue home: 3.29 vs 7.01 ERA (ok) · opp run factor 0.98 (weak) · 15 prior appearances this season
ATL	Didier Fuentes (Middle Reliever (High Leverage))	3 / 4	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0.32	1 / 0	1.72

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Elly De La Cruz (SS)	CIN	3 / 4	2 / 1	0 / 1	0 / 0	1 / 0	2 / 1	1 / 1	2 / 1	0 / 1	1.42	0.375 elite	projection too low
Matt Olson (1B)	ATL	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	0 / 1	-1.26	0.351 elite	bad day
Ronald Acuña Jr. (RF)	ATL	1 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	-1.11	0.368 elite	bad day
Eugenio Suárez (3B)	CIN	4 / 3	1 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 0	0 / 0	0 / 1	0.93	0.298 below average	good day
Michael Harris II (CF)	ATL	4 / 4	2 / 1	1 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	0 / 1	0.78	0.339 above average	projection too low
Sal Stewart (1B)	CIN	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 0	1 / 1	-0.6	0.349 elite	bad day
Dane Myers (CF)	CIN	4 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 1	1 / 1	0.6	0.31 average	no signal
Héctor Rodríguez (DH)	CIN	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	-0.47	0.249 weak	—
Matt McLain (2B)	CIN	4 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	1 / 1	0.47	0.298 below average	—
Drake Baldwin (DH)	ATL	3 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	0.47	0.367 elite	—
Ozzie Albies (2B)	ATL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	-0.47	0.279 weak	—
Sean Murphy (C)	ATL	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	-0.47	0.265 weak	—
JJ Bleday (LF)	CIN	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0.14	0.34 elite	—
Tyler Stephenson (C)	CIN	4 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	2 / 1	-0.02	0.328 above average	—
Carlos Jorge (RF)	CIN	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	3 / 1	0	0.247 weak	—
Austin Riley (3B)	ATL	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.297 below average	—
Mauricio Dubón (LF)	ATL	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0	0.306 average	—
Ha-Seong Kim (SS)	ATL	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.215 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	1.699	positive favors home; negative favors away
Recent form	1.245	positive favors home; negative favors away
Starting-pitcher matchup	0.059	positive favors home; negative favors away
Composite	1.917	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.91-4.94	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-8	0.6	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, matchup_recency_guardrail)
embedded_savant_run_environment	6-9	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-9	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-9	n/a	Capped at max 12 realistic runs
simulation_result	6-9	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.91-4.94	4-5	-0.021	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	5-8	2	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, matchup_recency_guardrail)
embedded_savant_run_environment	5-8	6-9	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	6-9	6-9	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	6-9	6-9	0	Capped at max 12 realistic runs
simulation_result	6-9	6-9	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
ATL	Drake Baldwin	batter / DH	better	predicted_RBI 5.417
CIN	Carlos Jorge	batter / RF	better	predicted_AB 2.5
CIN	Brandon Williamson	pitcher / Starter	worse	predicted_H 2
CIN	Elly De La Cruz	batter / SS	worse	predicted_SO 1.799
CIN	Matt McLain	batter / 2B	better	predicted_RBI 1.74
CIN	Dane Myers	batter / CF	worse	predicted_AB 1.67
ATL	Michael Harris II	batter / CF	worse	predicted_AB -1.66
CIN	Eugenio Suárez	batter / 3B	worse	predicted_SO 1.653

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	ATL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	ATL
info	CURRENT_PITCHER_DUPLICATION	CIN pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	CIN
info	CURRENT_FINALIZED_MATCH_BOX	ATL model-prediction batting totals differ from final simulation box score (H: current=8, sim=9; HR: current=1, sim=3; RBI: current=4, sim=9; R: current=5, sim=9; BB: current=3, sim=6; SO: current=7, sim=6; HBP: current=0, sim=1).	ATL
info	CURRENT_PLAYER_MATCH_BOX	ATL current finalized hitter lines are model predictions and differ from the simulation box score: Drake Baldwin: H current=1 box=2, HR current=0 box=2, RBI current=1 box=6, SO current=1 box=0; Matt Olson: HR current=1 box=0, RBI current=1 box=0; Ozzie Albies: AB current=4 box=5, H current=1 box=2, HR current=0 box=1, RBI current=0 box=2; Ronald Acuña Jr.: 2B current=1 box=0, SO current=1 box=0; Michael Harris II: AB current=4 box=3, RBI current=1 box=0, BB current=0 box=1, SO current=1 box=0; Austin Riley: AB current=3 box=4, H current=1 box=0, R current=0 box=1, SO current=1 box=2; Mauricio Dubón: AB current=4 box=3, 2B current=0 box=1, R current=0 box=1, BB current=0 box=1; Sean Murphy: R current=0 box=1, BB current=0 box=1, SO current=1 box=0	ATL
info	CURRENT_FINALIZED_MATCH_BOX	CIN model pitcher projections differ from final simulation pitching box score (ER: current=8, sim=9).	CIN
info	CURRENT_FINALIZED_MATCH_BOX	CIN model-prediction batting totals differ from final simulation box score (AB: current=33, sim=37; H: current=8, sim=10; 2B: current=1, sim=2; HR: current=1, sim=3; RBI: current=4, sim=6; R: current=4, sim=6; BB: current=4, sim=5; SO: current=9, sim=12).	CIN
info	CURRENT_PLAYER_MATCH_BOX	CIN current finalized hitter lines are model predictions and differ from the simulation box score: Carlos Jorge: AB current=4 box=5, RBI current=0 box=1, R current=0 box=1; Sal Stewart: 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Eugenio Suárez: AB current=3 box=5, HR current=0 box=1, R current=0 box=1, SO current=1 box=3; JJ Bleday: HR current=0 box=1, RBI current=0 box=2, BB current=1 box=2; Elly De La Cruz: 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Héctor Rodríguez: AB current=4 box=3, BB current=0 box=1, SO current=1 box=0; Tyler Stephenson: H current=1 box=2, 2B current=0 box=1, RBI current=1 box=0, R current=0 box=1; Matt McLain: AB current=3 box=4, H current=0 box=1, HR current=0 box=1, RBI current=0 box=2	CIN
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

CLE@BOSMISS

Predicted: BOS 3-4 (conf +0.01, toss_up)Actual: CLE 3-2Brier 0.255total bias +2

Root cause: Genuine toss-up that flipped

Evidence-backed diagnosis

published system edge 0.01 (toss_up); this is not the half-scale distance from 50%
team-run absolute error 2; total bias +2
pregame home-minus-away evidence: season +0.127, recent -0.503, pitcher +0.027, composite +0.004
confidence pipeline reached 0.507 before later publication/recency controls; published home probability was 0.505
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

No large single-factor errors.

Team	Starter	Projected	Actual	ER err
CLE	Parker Messick	3 ER / 18 outs	1 ER / 15 outs	-2
BOS	Payton Tolle	3 ER / 17 outs	2 ER / 17 outs	-1

Decision internals: run factors home×0.987 / away×0.96; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Jahmai Jones (BOS DH): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Adley Rutschman (BOS C): 3-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 0.92]
Roman Anthony (BOS LF): 0-for-5, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
José Ramírez (CLE 3B): 1-for-5, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.64]

Pitcher versus batter: what actually broke this game

AWAY pitching

Parker Messick (CLE) threw 15 outs, 1 ER = 1.8 ERA against a projected 4.5 ERA (-2.7 ERA error)

Bullpen behind him: 1 ER over 6 outs.

The HOME lineup he faced produced 1 runs on 7 hits and 0 HR, against a projected 4 runs / 8 hits (-3 runs vs projection).

Who did the damage to him:

Adley Rutschman: 3-for-4, 0 HR, 0 RBI [+0.92 runs vs projection] season xwOBA 0.327, above average - real projection error
Nate Eaton: 1-for-3, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.289, below average - a good day, not a model error
Nick Sogard: 2-for-5, 0 HR, 0 RBI [+0.45 runs vs projection] season xwOBA 0.31, average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Payton Tolle (BOS) threw 17 outs, 2 ER = 3.18 ERA against a projected 4.76 ERA (-1.58 ERA error)

Bullpen behind him: 1 ER over 3 outs.

The AWAY lineup he faced produced 3 runs on 11 hits and 0 HR, against a projected 4 runs / 8 hits (-1 runs vs projection).

Who did the damage to him:

David Fry: 1-for-3, 0 HR, 1 RBI [+0.47 runs vs projection] season xwOBA 0.264, weak - a good day, not a model error
Chase DeLauter: 2-for-4, 0 HR, 0 RBI [+0.45 runs vs projection] season xwOBA 0.346, elite - a good day, not a model error
Angel Martínez: 3-for-4, 0 HR, 0 RBI [+0.32 runs vs projection] season xwOBA 0.304, average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
CLE	Parker Messick (Starter)	15 / 18	1.8 / 4.5	1 / 3	5 / 1	0 / 1	1 / 2	1.48	5 / 8	5.54	venue away: 2.21 vs 2.97 ERA (ok) · opp run factor 0.99 (average) · 30 prior appearances this season
CLE	Cade Smith (Closer)	3 / 4	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0.3	1 / 0	1.54
CLE	Hunter Gaddis (Setup Man)	3 / 2	9 / 13.5 (-4.50)	1 / 1	2 / 2	1 / 0	0 / 1	0.27	2 / 2	0.78
BOS	Payton Tolle (Starter)	17 / 17	3.18 / 4.76	2 / 3	8 / 4	0 / 2	2 / 2	1.56	5 / 6	6.84	venue home: 3.55 vs 2.5 ERA (worse here) · opp run factor 0.96 (average) · 25 prior appearances this season
BOS	Erik Miller (Setup Man)	3 / 3	9 / 0 (+9.00)	1 / 0	2 / 0	0 / 0	1 / 0	0.85	0 / 0	1.37

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Jahmai Jones (DH)	BOS	3 / 3	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	0 / 0	1 / 1	-1.4	0.33 above average	bad day
Adley Rutschman (C)	BOS	4 / 4	3 / 1	2 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0 / 1	0.92	0.327 above average	projection too low
Roman Anthony (LF)	BOS	5 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	2 / 1	-0.8	0.346 elite	bad day
José Ramírez (3B)	CLE	5 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.64	0.329 above average	bad day
David Fry (1B)	CLE	3 / 3	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	0 / 0	0 / 1	0.47	0.264 weak	—
Nate Eaton (CF)	BOS	3 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0.47	0.289 below average	—
Chase DeLauter (RF)	CLE	4 / 4	2 / 1	1 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 1	0 / 1	0.45	0.346 elite	—
Nick Sogard (1B)	BOS	5 / 4	2 / 1	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	2 / 1	0.45	0.31 average	—
Andruw Monasterio (2B)	BOS	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	1 / 1	0 / 0	1 / 0	2 / 1	-0.45	0.292 below average	—
Angel Martínez (LF)	CLE	4 / 4	3 / 1	1 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 0	0 / 1	0.32	0.304 average	—
Wilyer Abreu (RF)	BOS	4 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	0.31	0.35 elite	—
Austin Hedges (C)	CLE	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 0	1 / 1	0.19	0.277 weak	—
Isiah Kiner-Falefa (SS)	BOS	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	2 / 1	-0.14	0.297 below average	—
Caleb Durbin (3B)	BOS	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	-0.14	0.295 below average	—
Jo Adell (DH)	CLE	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	0.02	0.325 above average	—
Steven Kwan (CF)	CLE	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	2 / 0	0	0.327 above average	—
Travis Bazzana (2B)	CLE	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	3 / 1	0	0.303 average	—
Brayan Rocchio (SS)	CLE	5 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0	0.295 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.127	positive favors home; negative favors away
Recent form	-0.503	positive favors home; negative favors away
Starting-pitcher matchup	0.027	positive favors home; negative favors away
Composite	0.004	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.13-4.82	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-3	0.505	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	3-3	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-4	n/a	Capped at max 12 realistic runs
simulation_result	3-4	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.13-4.82	4-5	0.312	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	4-3	-2	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-3	3-3	1	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-3	3-4	1	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-4	3-4	0	Capped at max 12 realistic runs
simulation_result	3-4	3-4	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
BOS	Payton Tolle	pitcher / Starter	neutral	predicted_SO 5
CLE	Parker Messick	pitcher / Starter	better	predicted_SO 2
BOS	Erik Miller	pitcher / Setup Man	worse	predicted_SO -2
BOS	Jahmai Jones	batter / DH	worse	predicted_RBI 1.92
CLE	Travis Bazzana	batter / 2B	worse	predicted_AB -1.78
CLE	Austin Hedges	batter / C	better	predicted_RBI 1.72
BOS	Adley Rutschman	batter / C	worse	predicted_AB -1.7
BOS	Caleb Durbin	batter / 3B	worse	predicted_SO 1.465

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	BOS pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	BOS
info	CURRENT_PITCHER_DUPLICATION	CLE pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	CLE
info	CURRENT_FINALIZED_MATCH_BOX	BOS model-prediction batting totals differ from final simulation box score (AB: current=32, sim=30; H: current=8, sim=3; 2B: current=2, sim=0; SO: current=8, sim=10; HBP: current=0, sim=1).	BOS
info	CURRENT_PLAYER_MATCH_BOX	BOS current finalized hitter lines are model predictions and differ from the simulation box score: Nick Sogard: AB current=4 box=3, RBI current=0 box=1, R current=0 box=1, BB current=1 box=0; Adley Rutschman: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Roman Anthony: R current=0 box=1, BB current=1 box=0; Andruw Monasterio: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Isiah Kiner-Falefa: H current=1 box=0; Nate Eaton: AB current=3 box=2, BB current=0 box=1; Jahmai Jones: AB current=3 box=4, RBI current=0 box=2, SO current=1 box=2; Wilyer Abreu: H current=1 box=0, RBI current=1 box=0, R current=1 box=0	BOS
info	CURRENT_FINALIZED_MATCH_BOX	CLE model-prediction batting totals differ from final simulation box score (AB: current=32, sim=31; H: current=8, sim=4; 2B: current=2, sim=1; HR: current=1, sim=2; RBI: current=4, sim=3; R: current=4, sim=3; BB: current=3, sim=2; SO: current=7, sim=6).	CLE
info	CURRENT_PLAYER_MATCH_BOX	CLE current finalized hitter lines are model predictions and differ from the simulation box score: Steven Kwan: H current=1 box=0, R current=1 box=0, BB current=1 box=0, SO current=0 box=1; Jo Adell: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, SO current=1 box=2; David Fry: H current=0 box=1, 2B current=0 box=1, R current=0 box=1, BB current=0 box=1; Angel Martínez: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; José Ramírez: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; Travis Bazzana: AB current=3 box=2, R current=1 box=0, BB current=0 box=1; Chase DeLauter: HR current=0 box=1, R current=0 box=1, BB current=1 box=0, SO current=1 box=0; Austin Hedges: HR current=0 box=1, RBI current=0 box=2, R current=0 box=1, SO current=1 box=0	CLE
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

HOU@SEAMISS

Predicted: SEA 3-5 (conf +0.20, strong)Actual: HOU 7-0Brier 0.36total bias +1

Root cause: Backed the wrong side with real conviction

Evidence-backed diagnosis

published system edge 0.20 (strong); this is not the half-scale distance from 50%
team-run absolute error 9; total bias +1
pregame home-minus-away evidence: season +0.254, recent +0.830, pitcher +0.078, composite +0.878
confidence pipeline reached 0.641 before later publication/recency controls; published home probability was 0.600
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

HOU offense under-projected by 4
SEA offense over-projected by 5
Cristian Javier (HOU) allowed 3 fewer ER than projected (0 vs 3)
committed to SEA at 0.20 confidence and it lost

Team	Starter	Projected	Actual	ER err
HOU	Cristian Javier	3 ER / 15 outs	0 ER / 6 outs	-3
SEA	Logan Gilbert	3 ER / 17 outs	4 ER / 12 outs	1

Decision internals: run factors home×1.056 / away×0.978; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Isaac Paredes (HOU 3B): 2-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.4]
Cal Raleigh (SEA C): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Christian Walker (HOU 1B): 1-for-5, 1 HR, 3 RBI (projected 1.0 H, 0.0 HR) [run impact 0.93]
Brice Matthews (HOU CF): 1-for-2, 0 HR, 0 RBI (projected 0.0 H, 0.0 HR) [run impact 0.8]
Cam Smith (HOU RF): 1-for-5, 0 HR, 0 RBI (projected 0.0 H, 0.0 HR) [run impact 0.78]

Starter miss: Cristian Javier (HOU) posted 0.00 ERA vs projected 5.40 (0 ER in 6 outs); he is materially worse pitching away (7.23 ERA there vs 5.28 on the other side), a venue-role weakness the model under-weighted.

What to do: Enable/raise the venue-split weight (context.venue_adjustment_weight) and validate in shadow; the split is credible for this starter.

Pitcher versus batter: what actually broke this game

AWAY pitching

Cristian Javier (HOU) threw 6 outs, 0 ER = 0 ERA against a projected 5.4 ERA (-5.4 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The HOME lineup he faced produced 0 runs on 4 hits and 0 HR, against a projected 4 runs / 8 hits (-4 runs vs projection).

Who did the damage to him:

Julio Rodríguez: 2-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.333, above average - a good day, not a model error
Cole Young: 1-for-3, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.314, average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Logan Gilbert (SEA) threw 12 outs, 4 ER = 9 ERA against a projected 4.76 ERA (4.24 ERA error)

Bullpen behind him: 1 ER over 5 outs.

The AWAY lineup he faced produced 7 runs on 12 hits and 3 HR, against a projected 4 runs / 7 hits (3 runs vs projection).

Who did the damage to him:

Isaac Paredes: 2-for-4, 1 HR, 1 RBI [+1.4 runs vs projection] season xwOBA 0.321, above average - real projection error
Christian Walker: 1-for-5, 1 HR, 3 RBI [+0.93 runs vs projection] season xwOBA 0.298, below average - a good day, not a model error
Brice Matthews: 1-for-2, 0 HR, 0 RBI [+0.8 runs vs projection] season xwOBA 0.257, weak - a good day, not a model error

Verdict: both: the starter was hit hard AND the lineup underproduced

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
HOU	Cristian Javier (Starter)	6 / 15	0 / 5.4 (-5.40)	0 / 3	1 / 6	0 / 1	2 / 3	1.93	2 / 2	4.8	venue away: 7.23 vs 5.28 ERA (worse here) · opp run factor 1.06 (weak) · 16 prior appearances this season
HOU	Logan VanWey (Setup Man)	6 / 4	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0	2 / 2	0.77
SEA	Logan Gilbert (Starter)	12 / 17	9 / 4.76 (+4.24)	4 / 3	10 / 6	1 / 2	2 / 1	1.78	3 / 3	7.2	venue home: 4.41 vs 2.72 ERA (worse here) · opp run factor 0.98 (average) · 30 prior appearances this season
SEA	José A. Ferrer (Setup Man)	5 / 4	5.4 / 0 (+5.40)	1 / 0	1 / 0	1 / 0	0 / 1	0.21	2 / 0	0.62

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Isaac Paredes (3B)	HOU	4 / 4	2 / 1	0 / 0	0 / 0	1 / 0	1 / 1	1 / 1	1 / 1	0 / 1	1.4	0.321 above average	projection too low
Cal Raleigh (C)	SEA	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	0 / 1	-1.4	0.312 average	no signal
Christian Walker (1B)	HOU	5 / 3	1 / 1	0 / 0	0 / 0	1 / 0	3 / 1	1 / 0	0 / 0	1 / 1	0.93	0.298 below average	good day
Brice Matthews (CF)	HOU	2 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 2	0.8	0.257 weak	good day
Cam Smith (RF)	HOU	5 / 3	1 / 0	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0.78	0.315 average	no signal
Yordan Alvarez (DH)	HOU	4 / 4	2 / 1	0 / 0	0 / 0	1 / 1	3 / 1	1 / 1	1 / 1	0 / 1	0.47	0.442 elite	—
Yainer Diaz (C)	HOU	5 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	1 / 1	0.47	0.304 average	—
J.P. Crawford (SS)	SEA	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	-0.47	0.322 above average	—
Lazaro Montes (RF)	SEA	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	-0.47	0.351 elite	—
Julio Rodríguez (CF)	SEA	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0.47	0.333 above average	—
Josh Naylor (1B)	SEA	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.312 average	—
Cole Young (2B)	SEA	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	2 / 1	0.33	0.314 average	—
Dominic Canzone (DH)	SEA	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	-0.31	0.357 elite	—
Taylor Trammell (LF)	HOU	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 0	0 / 1	0.19	0.29 below average	—
Jose Altuve (2B)	HOU	5 / 4	2 / 1	0 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	2 / 1	0.16	0.273 weak	—
Randy Arozarena (LF)	SEA	1 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	1 / 1	-0.14	0.348 elite	—
Jeremy Peña (SS)	HOU	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0 / 0	2 / 1	0	0.326 above average	—
Brock Rodden (3B)	SEA	1 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.236 weak	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.254	positive favors home; negative favors away
Recent form	0.83	positive favors home; negative favors away
Starting-pitcher matchup	0.078	positive favors home; negative favors away
Composite	0.878	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.85-4.98	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	3-5	0.6	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	3-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-5	n/a	Capped at max 12 realistic runs
simulation_result	3-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.85-4.98	4-5	-0.13	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	3-5	1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	3-5	3-5	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-5	3-5	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-5	3-5	0	Capped at max 12 realistic runs
simulation_result	3-5	3-5	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
SEA	Brock Rodden	batter / 3B	worse	predicted_BB 1.609
SEA	Cole Young	batter / 2B	better	predicted_RBI 1.6
SEA	Dominic Canzone	batter / DH	worse	predicted_AB -1.44
SEA	J.P. Crawford	batter / SS	worse	predicted_BB 1.436
SEA	Josh Naylor	batter / 1B	worse	predicted_AB -1.42
SEA	Lazaro Montes	batter / RF	worse	predicted_SO -1.357
HOU	Jose Altuve	batter / 2B	worse	predicted_H -1.24
SEA	Cal Raleigh	batter / C	worse	predicted_AB -1.24

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	HOU pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	HOU
info	CURRENT_PITCHER_DUPLICATION	SEA pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	SEA
info	CURRENT_FINALIZED_MATCH_BOX	HOU model-prediction batting totals differ from final simulation box score (AB: current=32, sim=33; H: current=7, sim=6; 2B: current=1, sim=0; HR: current=1, sim=2; RBI: current=4, sim=3; R: current=4, sim=3; SO: current=10, sim=3; HBP: current=0, sim=1).	HOU
info	CURRENT_PLAYER_MATCH_BOX	HOU current finalized hitter lines are model predictions and differ from the simulation box score: Jose Altuve: H current=1 box=0, 2B current=1 box=0, R current=1 box=0; Jeremy Peña: H current=1 box=2, RBI current=0 box=1, SO current=1 box=0; Isaac Paredes: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Yordan Alvarez: HR current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Yainer Diaz: HR current=0 box=1, R current=0 box=1, SO current=1 box=0; Taylor Trammell: H current=1 box=0, BB current=0 box=1, SO current=1 box=0; Cam Smith: H current=0 box=1, BB current=0 box=1, SO current=1 box=0; Christian Walker: HR current=0 box=1, R current=0 box=1, SO current=1 box=0, HBP current=0 box=1	HOU
info	CURRENT_FINALIZED_MATCH_BOX	SEA model-prediction batting totals differ from final simulation box score (AB: current=34, sim=31; H: current=8, sim=9; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=5; SO: current=9, sim=5; SF: current=0, sim=2).	SEA
info	CURRENT_PLAYER_MATCH_BOX	SEA current finalized hitter lines are model predictions and differ from the simulation box score: Randy Arozarena: AB current=4 box=5, 2B current=0 box=1, RBI current=1 box=0, R current=1 box=0; J.P. Crawford: AB current=4 box=3, H current=1 box=2, BB current=1 box=2; Dominic Canzone: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, SF current=0 box=1; Cal Raleigh: AB current=4 box=3, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Lazaro Montes: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Brock Rodden: AB current=3 box=2, BB current=0 box=2, SO current=1 box=0; Julio Rodríguez: H current=1 box=2, RBI current=0 box=1, R current=0 box=1, SO current=1 box=0; Josh Naylor: AB current=4 box=3, RBI current=0 box=1, R current=0 box=1, SF current=0 box=1	SEA
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

LAA@OAKMISS

Predicted: LAA 5-4 (conf -0.06, lean)Actual: OAK 7-9Brier 0.281total bias -7

Root cause: Run environment far higher than projected (under-dispersed)

Evidence-backed diagnosis

published system edge 0.06 (lean); this is not the half-scale distance from 50%
team-run absolute error 7; total bias -7
pregame home-minus-away evidence: season -0.684, recent -0.562, pitcher +0.128, composite -0.416
confidence pipeline reached 0.560 before later publication/recency controls; published home probability was 0.470
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs under-projected by 7 (9 vs 16)
OAK offense under-projected by 5

Team	Starter	Projected	Actual	ER err
LAA	Yusei Kikuchi	4 ER / 15 outs	2 ER / 16 outs	-2
OAK	Brady Basso	5 ER / 13 outs	5 ER / 6 outs	0

Decision internals: run factors home×1.089 / away×0.962; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Lawrence Butler (OAK RF): 3-for-4, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.54]
Denzer Guzman (LAA 3B): 1-for-4, 1 HR, 4 RBI (projected 1.0 H, 0.0 HR) [run impact 1.26]
Zack Gelof (OAK 3B): 1-for-3, 1 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.26]
Mike Trout (LAA DH): 3-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact 0.94]
Zach Neto (LAA SS): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]

Starter miss: Brady Basso (OAK) posted 22.50 ERA vs projected 10.38 (5 ER in 6 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

What to do: Variance — no model change is warranted for this start.

Pitcher versus batter: what actually broke this game

AWAY pitching

Yusei Kikuchi (LAA) threw 16 outs, 2 ER = 3.38 ERA against a projected 7.2 ERA (-3.82 ERA error)

Bullpen behind him: 3 ER over 1 outs.

The HOME lineup he faced produced 7 runs on 12 hits and 1 HR, against a projected 4 runs / 9 hits (3 runs vs projection).

Who did the damage to him:

Lawrence Butler: 3-for-4, 0 HR, 1 RBI [+1.54 runs vs projection] season xwOBA 0.324, above average - real projection error
Zack Gelof: 1-for-3, 1 HR, 2 RBI [+1.26 runs vs projection] season xwOBA 0.318, average - a good day, not a model error
Tommy White: 2-for-2, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.29, below average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Brady Basso (OAK) threw 6 outs, 5 ER = 22.5 ERA against a projected 10.38 ERA (12.12 ERA error)

Bullpen behind him: 2 ER over 14 outs.

The AWAY lineup he faced produced 7 runs on 15 hits and 1 HR, against a projected 3 runs / 9 hits (4 runs vs projection).

Who did the damage to him:

Denzer Guzman: 1-for-4, 1 HR, 4 RBI [+1.26 runs vs projection] season xwOBA 0.269, weak - a good day, not a model error
Mike Trout: 3-for-4, 0 HR, 0 RBI [+0.94 runs vs projection] season xwOBA 0.378, elite - real projection error
Vaughn Grissom: 3-for-4, 0 HR, 2 RBI [+0.63 runs vs projection] season xwOBA 0.308, average - a good day, not a model error

Verdict: pitching: the starter gave up runs at a far worse rate than projected

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
LAA	Yusei Kikuchi (Starter)	16 / 15	3.38 / 7.2 (-3.82)	2 / 4	7 / 5	0 / 0	1 / 2	2.57	2 / 6	5.64	venue away: 5.66 vs 4.03 ERA (worse here) · opp run factor 1.09 (weak) · 12 prior appearances this season
LAA	José Fermin (Middle Reliever (High Leverage))	0 / 3	n/a / 0	3 / 0	3 / 0	0 / 0	1 / 0	0.51	0 / 0	1.3
LAA	Tayler Saucedo (Middle Reliever (High Leverage))	1 / 2	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0.16	0 / 1	0.92
OAK	Brady Basso (Starter)	6 / 13	22.5 / 10.38 (+12.12)	5 / 5	6 / 5	1 / 2	2 / 2	0.62	2 / 1	2.38	venue home: 0 vs 7.59 ERA (ok) · opp run factor 0.96 (weak) · 24 prior appearances this season
OAK	Seth Johnson (Reliever)	8 / 5	3.38 / 0 (+3.38)	1 / 0	4 / 0	0 / 0	1 / 0	n/a	1 / 1	n/a
OAK	Hogan Harris (Reliever)	3 / 3	9 / 0 (+9.00)	1 / 0	3 / 0	0 / 0	1 / 0	n/a	1 / 0	n/a
OAK	Luis Medina (Reliever)	3 / 3	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	n/a	1 / 1	n/a

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Lawrence Butler (RF)	OAK	4 / 4	3 / 1	1 / 0	1 / 0	0 / 0	1 / 1	2 / 0	0 / 1	0 / 1	1.54	0.324 above average	projection too low
Denzer Guzman (3B)	LAA	4 / 4	1 / 1	0 / 0	0 / 0	1 / 0	4 / 1	1 / 0	1 / 0	0 / 1	1.26	0.269 weak	good day
Zack Gelof (3B)	OAK	3 / 4	1 / 1	0 / 0	0 / 0	1 / 0	2 / 1	1 / 1	1 / 0	0 / 1	1.26	0.318 average	no signal
Mike Trout (DH)	LAA	4 / 4	3 / 1	1 / 1	0 / 0	0 / 0	0 / 0	2 / 1	1 / 1	1 / 1	0.94	0.378 elite	projection too low
Zach Neto (SS)	LAA	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	2 / 1	1 / 1	0 / 1	-0.93	0.311 average	no signal
Joshua Kuroda-Grauer (2B)	OAK	2 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 0	-0.78	0.327 above average	bad day
Vaughn Grissom (1B)	LAA	4 / 4	3 / 1	0 / 1	0 / 0	0 / 0	2 / 1	1 / 0	0 / 0	0 / 0	0.63	0.308 average	no signal
Jose Siri (CF)	LAA	5 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	1 / 1	0.47	0.299 below average	—
Wade Meckler (RF)	LAA	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0.47	0.309 average	—
Christian Moore (2B)	LAA	5 / 3	2 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	0.47	0.225 weak	—
Bryce Teodosio (LF)	LAA	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	-0.47	0.202 weak	—
Tommy White (1B)	OAK	2 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0.47	0.29 below average	—
Denzel Clarke (CF)	OAK	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.214 weak	—
Jonah Heim (C)	OAK	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	-0.47	0.284 below average	—
Moisés Ballesteros (C)	LAA	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	0 / 1	0.33	0.293 below average	—
Henry Bolte (LF)	OAK	4 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	1 / 1	1 / 1	1 / 1	0.31	0.317 average	—
Alika Williams (SS)	OAK	4 / 3	2 / 1	0 / 1	0 / 0	0 / 0	0 / 0	2 / 0	0 / 0	0 / 1	0.16	0.24 weak	—
Shea Langeliers (DH)	OAK	5 / 4	3 / 1	1 / 0	0 / 0	0 / 1	1 / 1	1 / 1	0 / 1	0 / 1	-0.01	0.359 elite	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.684	positive favors home; negative favors away
Recent form	-0.562	positive favors home; negative favors away
Starting-pitcher matchup	0.128	positive favors home; negative favors away
Composite	-0.416	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	5.17-5.21	n/a	Base ML historical model prediction
variance_applied	5-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	0.53	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	n/a	Capped at max 12 realistic runs
simulation_result	5-4	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	5.17-5.21	5-5	0.04	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-5	5-4	1	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by winner_starter_is_below_average, projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	5-4	5-4	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-4	5-4	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-4	5-4	-0	Capped at max 12 realistic runs
simulation_result	5-4	5-4	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
OAK	Brady Basso	pitcher / Starter	worse	predicted_IP -4.9
OAK	Chris Roycroft	pitcher / Reliever	better	predicted_SO 2
OAK	Zack Gelof	batter / 3B	worse	predicted_AB -1.73
OAK	Jonah Heim	batter / C	worse	predicted_RBI 1.67
LAA	Jose Siri	batter / CF	better	predicted_RBI 1.643
OAK	Joshua Kuroda-Grauer	batter / 2B	worse	predicted_H -1.562
LAA	Bryce Teodosio	batter / LF	worse	predicted_AB 1.5
OAK	Denzel Clarke	batter / CF	worse	predicted_AB 1.444

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	LAA pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	LAA
info	CURRENT_PITCHER_DUPLICATION	OAK pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	OAK
info	CURRENT_FINALIZED_MATCH_BOX	LAA model-prediction batting totals differ from final simulation box score (H: current=9, sim=5; 2B: current=2, sim=1; HR: current=1, sim=2; R: current=3, sim=5; SO: current=8, sim=5; HBP: current=0, sim=1).	LAA
info	CURRENT_PLAYER_MATCH_BOX	LAA current finalized hitter lines are model predictions and differ from the simulation box score: Mike Trout: AB current=4 box=3, 2B current=1 box=0, BB current=1 box=0, SO current=1 box=0; Vaughn Grissom: 2B current=1 box=0, RBI current=1 box=2, R current=0 box=1; Jose Siri: AB current=3 box=4, HR current=0 box=1, RBI current=1 box=2, R current=0 box=1; Moisés Ballesteros: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Wade Meckler: H current=1 box=0, R current=1 box=0, SO current=1 box=2; Christian Moore: AB current=3 box=4, H current=1 box=0, SO current=1 box=0; Zach Neto: AB current=4 box=3; Denzer Guzman: H current=1 box=0, RBI current=1 box=0, SO current=1 box=0	LAA
info	CURRENT_FINALIZED_MATCH_BOX	OAK model-prediction batting totals differ from final simulation box score (AB: current=34, sim=33; H: current=9, sim=6; 2B: current=2, sim=3; HR: current=1, sim=0; RBI: current=5, sim=4; BB: current=3, sim=2; HBP: current=0, sim=1).	OAK
info	CURRENT_PLAYER_MATCH_BOX	OAK current finalized hitter lines are model predictions and differ from the simulation box score: Henry Bolte: 2B current=0 box=1, RBI current=0 box=1, BB current=1 box=0, SO current=1 box=0; Shea Langeliers: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Tommy White: 2B current=0 box=1; Joshua Kuroda-Grauer: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Alika Williams: AB current=3 box=4, 2B current=1 box=0, RBI current=0 box=1, R current=0 box=1; Zack Gelof: AB current=4 box=3, RBI current=1 box=0, HBP current=0 box=1; Denzel Clarke: AB current=3 box=4, H current=1 box=0, SO current=1 box=0; Jonah Heim: H current=1 box=2, 2B current=0 box=1, RBI current=1 box=2, R current=0 box=1	OAK
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

MIA@CHCMISS

Predicted: CHC 4-6 (conf +0.06, lean)Actual: MIA 8-2Brier 0.281total bias +0

Root cause: Starting-pitcher projection was wrong

Evidence-backed diagnosis

published system edge 0.06 (lean); this is not the half-scale distance from 50%
team-run absolute error 8; total bias +0
pregame home-minus-away evidence: season +0.936, recent +0.749, pitcher +0.087, composite +1.254
confidence pipeline reached 0.655 before later publication/recency controls; published home probability was 0.530
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

MIA offense under-projected by 4
CHC offense over-projected by 4
Janson Junk (MIA) allowed 5 fewer ER than projected (0 vs 5)

Team	Starter	Projected	Actual	ER err
MIA	Janson Junk	5 ER / 14 outs	0 ER / 15 outs	-5
CHC	Shota Imanaga	3 ER / 16 outs	1 ER / 15 outs	-2

Decision internals: run factors home×1.039 / away×0.952; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Kyle Stowers (MIA DH): 2-for-4, 1 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.71]
Heriberto Hernández (MIA LF): 1-for-5, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]
Pedro Ramirez (CHC 3B): 0-for-5, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.78]

Starter miss: Janson Junk (MIA) posted 0.00 ERA vs projected 9.64 (0 ER in 15 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

What to do: Variance — no model change is warranted for this start.

Pitcher versus batter: what actually broke this game

AWAY pitching

Janson Junk (MIA) threw 15 outs, 0 ER = 0 ERA against a projected 9.64 ERA (-9.64 ERA error)

Bullpen behind him: 0 ER over 9 outs.

The HOME lineup he faced produced 2 runs on 7 hits and 0 HR, against a projected 5 runs / 9 hits (-3 runs vs projection).

Who did the damage to him:

Michael Busch: 1-for-4, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.333, above average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Shota Imanaga (CHC) threw 15 outs, 1 ER = 1.8 ERA against a projected 5.06 ERA (-3.26 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The AWAY lineup he faced produced 7 runs on 8 hits and 1 HR, against a projected 5 runs / 8 hits (2 runs vs projection).

Who did the damage to him:

Kyle Stowers: 2-for-4, 1 HR, 2 RBI [+1.71 runs vs projection] season xwOBA 0.324, above average - real projection error
Javier Sanoja: 2-for-5, 0 HR, 1 RBI [+0.49 runs vs projection] season xwOBA 0.293, below average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
MIA	Janson Junk (Starter)	15 / 14	0 / 9.64 (-9.64)	0 / 5	2 / 6	0 / 0	1 / 4	1.2	6 / 5	3.07	venue away: 4.66 vs 4.46 ERA (ok) · opp run factor 1.04 (strong) · 24 prior appearances this season
MIA	Victor Vodnik (Middle Reliever (High Leverage))	3 / 4	0 / 6.75 (-6.75)	0 / 1	0 / 1	0 / 0	0 / 0	0.8	2 / 1	0.96
MIA	Calvin Faucher (Closer)	3 / 3	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	0.56	2 / 2	1.03
MIA	Michael Petersen (Middle Reliever (High Leverage))	3 / 3	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	0.37	1 / 1	1.48
CHC	Shota Imanaga (Starter)	15 / 16	1.8 / 5.06 (-3.26)	1 / 3	4 / 6	0 / 2	0 / 1	1.27	7 / 5	5.06	venue home: 4 vs 3.45 ERA (worse here) · opp run factor 0.95 (average) · 30 prior appearances this season
CHC	Daniel Palencia (Middle Reliever (High Leverage))	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0.39	1 / 0	0.97
CHC	Ryan Zeferjahn (Setup Man)	2 / 5	0 / 5.4 (-5.40)	0 / 1	1 / 1	0 / 1	1 / 0	0.65	1 / 0	1.39

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Kyle Stowers (DH)	MIA	4 / 4	2 / 1	1 / 0	0 / 0	1 / 0	2 / 1	3 / 0	1 / 1	1 / 1	1.71	0.324 above average	projection too low
Heriberto Hernández (LF)	MIA	5 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 1	1 / 1	-0.93	0.347 elite	bad day
Pedro Ramirez (3B)	CHC	5 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	-0.78	0.287 below average	projection too high
Javier Sanoja (3B)	MIA	5 / 4	2 / 1	0 / 1	0 / 0	0 / 0	1 / 1	0 / 0	1 / 0	1 / 0	0.49	0.293 below average	—
Seiya Suzuki (RF)	CHC	6 / 4	2 / 1	1 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	1 / 1	-0.48	0.339 above average	—
Esteury Ruiz (CF)	MIA	6 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	4 / 1	-0.47	0.282 weak	—
Deyvison De Los Santos (1B)	MIA	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.15 weak	—
Michael Conforto (DH)	CHC	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	-0.47	0.335 above average	—
Dansby Swanson (SS)	CHC	5 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	4 / 1	-0.47	0.308 average	—
Agustín Ramírez (C)	MIA	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 1	0 / 1	-0.33	0.285 below average	—
Michael Busch (1B)	CHC	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	2 / 1	1 / 1	0.33	0.333 above average	—
Ian Happ (LF)	CHC	6 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 1	2 / 1	-0.33	0.316 average	—
Pete Crow-Armstrong (CF)	CHC	4 / 4	2 / 1	1 / 0	0 / 0	0 / 1	0 / 1	1 / 1	2 / 1	1 / 1	0.18	0.377 elite	—
Nico Hoerner (2B)	CHC	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	-0.14	0.322 above average	—
Otto Lopez (SS)	MIA	5 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	1 / 1	0	0.328 above average	—
Jared Serna (2B)	MIA	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	2 / 1	0	0.235 weak	—
Griffin Conine (RF)	MIA	6 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	1 / 1	0 / 0	1 / 1	0	0.362 elite	—
Carson Kelly (C)	CHC	5 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	3 / 0	0	0.315 average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.936	positive favors home; negative favors away
Recent form	0.749	positive favors home; negative favors away
Starting-pitcher matchup	0.087	positive favors home; negative favors away
Composite	1.254	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.93-5.83	n/a	Base ML historical model prediction
variance_applied	4-6	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-6	0.53	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	4-6	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-6	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-6	n/a	Capped at max 12 realistic runs
simulation_result	4-6	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.93-5.83	4-6	0.101	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-6	4-6	0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	4-6	4-6	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-6	4-6	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-6	4-6	0	Capped at max 12 realistic runs
simulation_result	4-6	4-6	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
CHC	Pete Crow-Armstrong	batter / CF	worse	predicted_AB -3.43
CHC	Shota Imanaga	pitcher / Starter	better	predicted_SO 2
CHC	Carson Kelly	batter / C	better	predicted_2B 1.865
CHC	Nico Hoerner	batter / 2B	worse	predicted_AB -1.82
MIA	Heriberto Hernández	batter / LF	better	predicted_RBI 1.52
MIA	Esteury Ruiz	batter / CF	worse	predicted_AB 1.5
MIA	Deyvison De Los Santos	batter / 1B	better	predicted_AB 1.5
MIA	Jared Serna	batter / 2B	worse	predicted_AB 1.5

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	CHC pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	CHC
info	CURRENT_PITCHER_DUPLICATION	MIA pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	MIA
info	CURRENT_FINALIZED_MATCH_BOX	CHC model-prediction batting totals differ from final simulation box score (AB: current=33, sim=31; H: current=9, sim=8; 2B: current=1, sim=2; 3B: current=0, sim=1; HR: current=2, sim=0; RBI: current=5, sim=6; R: current=5, sim=6; BB: current=4, sim=6).	CHC
info	CURRENT_PLAYER_MATCH_BOX	CHC current finalized hitter lines are model predictions and differ from the simulation box score: Seiya Suzuki: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Michael Conforto: AB current=3 box=4, H current=1 box=2, RBI current=1 box=2, R current=0 box=1; Michael Busch: RBI current=0 box=2; Pedro Ramirez: AB current=4 box=3, H current=1 box=2, 2B current=1 box=0, RBI current=1 box=0; Pete Crow-Armstrong: AB current=4 box=3, H current=1 box=0, HR current=1 box=0, SO current=1 box=2; Nico Hoerner: AB current=4 box=3, H current=1 box=0, BB current=0 box=1, SO current=0 box=2; Ian Happ: H current=1 box=0, R current=1 box=0, BB current=1 box=0; Carson Kelly: H current=1 box=2, 2B current=0 box=2, R current=0 box=1, BB current=0 box=1	CHC
info	CURRENT_FINALIZED_MATCH_BOX	MIA model pitcher projections differ from final simulation pitching box score (BB: current=5, sim=6).	MIA
info	CURRENT_FINALIZED_MATCH_BOX	MIA model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; HR: current=1, sim=3; R: current=5, sim=4; BB: current=3, sim=1; SO: current=8, sim=5).	MIA
info	CURRENT_PLAYER_MATCH_BOX	MIA current finalized hitter lines are model predictions and differ from the simulation box score: Kyle Stowers: H current=1 box=2, HR current=0 box=1, R current=0 box=1, BB current=1 box=0; Agustín Ramírez: AB current=4 box=3, HR current=0 box=1, RBI current=0 box=1; Esteury Ruiz: AB current=3 box=4, 2B current=0 box=1; Otto Lopez: SO current=1 box=0; Deyvison De Los Santos: AB current=3 box=4, SO current=1 box=2; Jared Serna: AB current=3 box=4, R current=1 box=0, SO current=1 box=0; Heriberto Hernández: H current=1 box=2, RBI current=1 box=2, BB current=1 box=0, SO current=1 box=0; Javier Sanoja: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0	MIA
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

MIL@PHIMISS

Predicted: MIL 7-4 (conf -0.20, strong)Actual: PHI 4-6Brier 0.36total bias +1

Root cause: Backed the wrong side with real conviction

Evidence-backed diagnosis

published system edge 0.20 (strong); this is not the half-scale distance from 50%
team-run absolute error 5; total bias +1
pregame home-minus-away evidence: season -1.149, recent -1.125, pitcher +0.004, composite -1.444
confidence pipeline reached 0.677 before later publication/recency controls; published home probability was 0.400
score-volatility actions: supported_blowout_margin_expansion
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

MIL offense over-projected by 3
committed to MIL at 0.20 confidence and it lost

Team	Starter	Projected	Actual	ER err
MIL	Dustin May	2 ER / 14 outs	2 ER / 6 outs	0
PHI	Zack Wheeler	2 ER / 18 outs	3 ER / 17 outs	1

Decision internals: run factors home×0.976 / away×0.972; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Kyle Schwarber (PHI DH): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.73]
Brandon Marsh (PHI LF): 2-for-4, 1 HR, 3 RBI (projected 1.0 H, 0.0 HR) [run impact 1.71]
Jackson Chourio (MIL LF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.07]
Alec Bohm (PHI 1B): 2-for-4, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.78]
Garrett Mitchell (MIL CF): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.64]

Starter miss: Dustin May (MIL) posted 9.00 ERA vs projected 3.86 (2 ER in 6 outs); he is materially worse pitching away (5.76 ERA there vs 3.87 on the other side), a venue-role weakness the model under-weighted.

What to do: Enable/raise the venue-split weight (context.venue_adjustment_weight) and validate in shadow; the split is credible for this starter.

Pitcher versus batter: what actually broke this game

AWAY pitching

Dustin May (MIL) threw 6 outs, 2 ER = 9 ERA against a projected 3.86 ERA (5.14 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 6 runs on 8 hits and 1 HR, against a projected 4 runs / 8 hits (2 runs vs projection).

Who did the damage to him:

Brandon Marsh: 2-for-4, 1 HR, 3 RBI [+1.71 runs vs projection] season xwOBA 0.3, below average - a good day, not a model error
Alec Bohm: 2-for-4, 0 HR, 1 RBI [+0.78 runs vs projection] season xwOBA 0.33, above average - real projection error

Verdict: pitching: the starter gave up runs at a far worse rate than projected

HOME pitching

Zack Wheeler (PHI) threw 17 outs, 3 ER = 4.76 ERA against a projected 3 ERA (1.76 ERA error)

Bullpen behind him: 1 ER over 3 outs.

The AWAY lineup he faced produced 4 runs on 6 hits and 0 HR, against a projected 5 runs / 8 hits (-1 runs vs projection).

Who did the damage to him:

David Hamilton: 1-for-3, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.269, weak - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
MIL	Dustin May (Starter)	6 / 14	9 / 3.86 (+5.14)	2 / 2	1 / 2	1 / 0	2 / 2	1.66	2 / 4	5.22	venue away: 5.76 vs 3.87 ERA (worse here) · opp run factor 0.98 (average) · 29 prior appearances this season
PHI	Zack Wheeler (Starter)	17 / 18	4.76 / 3	3 / 2	5 / 5	0 / 0	1 / 3	2.49	11 / 7	5.83	venue home: 2.86 vs 3.1 ERA (ok) · opp run factor 0.97 (strong) · 26 prior appearances this season
PHI	Jhoan Duran (Closer)	3 / 3	9 / 0 (+9.00)	1 / 0	1 / 0	0 / 0	1 / 0	0.13	2 / 2	1.28

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Kyle Schwarber (DH)	PHI	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	1 / 1	-1.73	0.365 elite	bad day
Brandon Marsh (LF)	PHI	4 / 3	2 / 1	1 / 0	0 / 0	1 / 0	3 / 0	1 / 1	0 / 0	1 / 1	1.71	0.3 below average	good day
Jackson Chourio (LF)	MIL	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 0	2 / 2	-1.07	0.34 elite	bad day
Alec Bohm (1B)	PHI	4 / 4	2 / 1	1 / 0	0 / 0	0 / 0	1 / 1	1 / 0	0 / 0	0 / 1	0.78	0.33 above average	projection too low
Garrett Mitchell (CF)	MIL	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	2 / 2	-0.64	0.355 elite	bad day
Cooper Pratt (SS)	MIL	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.318 average	—
David Hamilton (3B)	MIL	3 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0.47	0.269 weak	—
Bryce Harper (RF)	PHI	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	-0.47	0.381 elite	—
Brice Turang (2B)	MIL	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	-0.33	0.344 elite	—
Jake Bauers (1B)	MIL	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	3 / 1	-0.33	0.38 elite	—
Luis Arraez (2B)	PHI	4 / 4	2 / 1	0 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	0.16	0.308 average	—
Christian Yelich (DH)	MIL	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 0	2 / 1	-0.14	0.308 average	—
J.T. Realmuto (C)	PHI	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	-0.14	0.32 average	—
Luis Lara (RF)	MIL	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	1 / 1	0	0.292 below average	—
William Contreras (C)	MIL	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	3 / 1	0 / 1	0 / 0	2 / 1	0	0.336 above average	—
Bryson Stott (3B)	PHI	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	2 / 0	1 / 1	0 / 1	0	0.327 above average	—
Trea Turner (SS)	PHI	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 1	0	0.295 below average	—
Justin Crawford (CF)	PHI	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.284 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-1.149	positive favors home; negative favors away
Recent form	-1.125	positive favors home; negative favors away
Starting-pitcher matchup	0.004	positive favors home; negative favors away
Composite	-1.444	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.88-5.44	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	7-4	0.6	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	7-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	7-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	7-4	n/a	Capped at max 12 realistic runs
simulation_result	7-4	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.88-5.44	4-5	0.564	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	7-4	4	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	7-4	7-4	-0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	7-4	7-4	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	7-4	7-4	-0	Capped at max 12 realistic runs
simulation_result	7-4	7-4	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
PHI	Luis Arraez	batter / 2B	worse	predicted_BB 2.767
MIL	Jackson Chourio	batter / LF	worse	predicted_RBI 2.38
PHI	Brandon Marsh	batter / LF	worse	predicted_SO 2.075
PHI	Zack Wheeler	pitcher / Starter	better	predicted_ER -2
MIL	Luis Lara	batter / RF	better	predicted_2B 1.863
MIL	William Contreras	batter / C	better	predicted_2B 1.814
MIL	Jake Bauers	batter / 1B	better	predicted_AB 1.636
PHI	Bryson Stott	batter / 3B	better	predicted_RBI 1.568

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	MIL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	MIL
info	CURRENT_PITCHER_DUPLICATION	PHI pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	PHI
info	CURRENT_FINALIZED_MATCH_BOX	MIL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=39; H: current=8, sim=9; 2B: current=1, sim=7; HR: current=1, sim=0; RBI: current=5, sim=7; R: current=5, sim=7; SO: current=11, sim=9).	MIL
info	CURRENT_PLAYER_MATCH_BOX	MIL current finalized hitter lines are model predictions and differ from the simulation box score: Garrett Mitchell: H current=1 box=0, 2B current=1 box=0, R current=0 box=1, SO current=2 box=1; Brice Turang: AB current=4 box=5, H current=1 box=2, 2B current=0 box=1, BB current=1 box=0; Jackson Chourio: AB current=4 box=5, 2B current=0 box=1, HR current=1 box=0, RBI current=1 box=3; Luis Lara: AB current=4 box=3, H current=1 box=2, 2B current=0 box=2, RBI current=0 box=1; William Contreras: AB current=4 box=5, H current=1 box=3, 2B current=0 box=2, RBI current=1 box=0; Jake Bauers: AB current=3 box=5, 2B current=0 box=1, BB current=1 box=0, SO current=1 box=0; Christian Yelich: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Cooper Pratt: H current=1 box=0, BB current=0 box=1	MIL
info	CURRENT_FINALIZED_MATCH_BOX	PHI model pitcher projections differ from final simulation pitching box score (ER: current=6, sim=7).	PHI
info	CURRENT_FINALIZED_MATCH_BOX	PHI model-prediction batting totals differ from final simulation box score (H: current=8, sim=3; 3B: current=0, sim=1; HR: current=1, sim=0; BB: current=3, sim=8; SO: current=8, sim=9; HBP: current=0, sim=1).	PHI
info	CURRENT_PLAYER_MATCH_BOX	PHI current finalized hitter lines are model predictions and differ from the simulation box score: Bryce Harper: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=1 box=0; Alec Bohm: H current=1 box=0, RBI current=1 box=0, HBP current=0 box=1; Bryson Stott: 3B current=0 box=1, RBI current=1 box=2, R current=0 box=1, SO current=1 box=0; Trea Turner: H current=1 box=0, R current=1 box=0, BB current=0 box=1; J.T. Realmuto: AB current=3 box=4, H current=1 box=0, SO current=1 box=2; Kyle Schwarber: HR current=1 box=0, SO current=1 box=0; Luis Arraez: AB current=4 box=2, R current=0 box=1, BB current=0 box=3; Brandon Marsh: H current=1 box=0, RBI current=0 box=1, BB current=0 box=1, SO current=1 box=3	PHI
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

MIN@SFHIT

Predicted: MIN 5-3 (conf -0.03, toss_up)Actual: MIN 3-2Brier 0.235total bias +3

Root cause: Toss-up that went our way

Evidence-backed diagnosis

published system edge 0.03 (toss_up); this is not the half-scale distance from 50%
team-run absolute error 3; total bias +3
pregame home-minus-away evidence: season -0.124, recent +0.302, pitcher -0.060, composite -0.484
confidence pipeline reached 0.523 before later publication/recency controls; published home probability was 0.485
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 3 (8 vs 5)

Team	Starter	Projected	Actual	ER err
MIN	Taj Bradley	3 ER / 18 outs	1 ER / 18 outs	-2
SF	Anthony Molina	3 ER / 15 outs	2 ER / 18 outs	-1

Decision internals: run factors home×0.949 / away×1.009; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Josh Bell (MIN DH): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Turner Hill (SF LF): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Victor Bericoto (SF 1B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.07]
Kody Clemens (MIN 2B): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.93]
Drew Cavanaugh (SF C): 0-for-1, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Starter miss: Taj Bradley (MIN) posted 1.50 ERA vs projected 4.50 (1 ER in 18 outs); he is materially worse pitching away (4.47 ERA there vs 3.05 on the other side), a venue-role weakness the model under-weighted.

What to do: Enable/raise the venue-split weight (context.venue_adjustment_weight) and validate in shadow; the split is credible for this starter.

Pitcher versus batter: what actually broke this game

AWAY pitching

Taj Bradley (MIN) threw 18 outs, 1 ER = 1.5 ERA against a projected 4.5 ERA (-3 ERA error)

Bullpen behind him: 1 ER over 5 outs.

The HOME lineup he faced produced 1 runs on 5 hits and 0 HR, against a projected 5 runs / 7 hits (-4 runs vs projection).

Who did the damage to him:

Scott Bandura: 1-for-2, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.113, weak - a good day, not a model error
Christian Koss: 1-for-3, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.25, weak - a good day, not a model error
Shay Whitcomb: 1-for-3, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.251, weak - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Anthony Molina (SF) threw 18 outs, 2 ER = 3 ERA against a projected 5.4 ERA (-2.4 ERA error) only 8 prior appearances in the covered season - the projection had little of his own record to stand on

Bullpen behind him: 0 ER over 3 outs.

The AWAY lineup he faced produced 3 runs on 6 hits and 1 HR, against a projected 4 runs / 8 hits (-1 runs vs projection).

Who did the damage to him:

Walker Jenkins: 1-for-4, 1 HR, 1 RBI [+0.6 runs vs projection] season xwOBA 0.31, average - a good day, not a model error
Brooks Lee: 2-for-4, 0 HR, 1 RBI [+0.47 runs vs projection] season xwOBA 0.278, weak - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
MIN	Taj Bradley (Starter)	18 / 18	1.5 / 4.5 (-3.00)	1 / 3	3 / 2	0 / 1	3 / 3	2.05	6 / 6	7.2	venue away: 4.47 vs 3.05 ERA (worse here) · opp run factor 0.95 (weak) · 30 prior appearances this season
MIN	A.J. Minter (Middle Reliever (High Leverage))	2 / 3	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0.17	1 / 0	1
MIN	Jeff Hoffman (Setup Man)	3 / 4	9 / 0 (+9.00)	1 / 0	0 / 0	0 / 0	2 / 0	0.44	1 / 1	1.56
SF	Anthony Molina (Starter)	18 / 15	3 / 5.4	2 / 3	4 / 4	0 / 0	1 / 2	0.88	6 / 7	6.35	venue home: 21 vs 1.72 ERA (ok) · opp run factor 1.01 (average) · 8 prior appearances this season
SF	Carson Seymour (Middle Reliever (High Leverage))	3 / 3	0 / 9 (-9.00)	0 / 1	0 / 1	0 / 1	0 / 0	0.52	0 / 1	0.87

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Josh Bell (DH)	MIN	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	1 / 1	-1.11	0.335 above average	bad day
Turner Hill (LF)	SF	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	0 / 1	-1.11	0.323 above average	bad day
Victor Bericoto (1B)	SF	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 0	2 / 1	-1.07	0.302 below average	projection too high
Kody Clemens (2B)	MIN	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-0.93	0.331 above average	bad day
Drew Cavanaugh (C)	SF	1 / 3	0 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0 / 1	-0.8	0.305 average	no signal
Walker Jenkins (CF)	MIN	4 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 1	0 / 1	0 / 1	0.6	0.31 average	no signal
Brooks Lee (3B)	MIN	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	0 / 0	1 / 0	0.47	0.278 weak	—
Royce Lewis (1B)	MIN	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.306 average	—
Scott Bandura (RF)	SF	2 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0.47	0.113 weak	—
Christian Koss (SS)	SF	3 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	0.47	0.25 weak	—
Shay Whitcomb (3B)	SF	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	1 / 1	0.33	0.251 weak	—
Luke Keaschall (RF)	MIN	4 / 4	1 / 1	0 / 0	1 / 0	0 / 0	1 / 0	1 / 1	0 / 1	0 / 0	0.29	0.315 average	—
Emmanuel Rodriguez (LF)	MIN	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	-0.14	0.354 elite	—
Jonah Cox (2B)	SF	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 1	1 / 0	2 / 2	-0.14	0.264 weak	—
Victor Caratini (C)	MIN	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.309 average	—
Ryan Kreidler (SS)	MIN	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	0	0.295 below average	—
Bo Davidson (DH)	SF	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	0	0.309 average	—
Drew Gilbert (CF)	SF	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	0	0.333 above average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	-0.124	positive favors home; negative favors away
Recent form	0.302	positive favors home; negative favors away
Starting-pitcher matchup	-0.06	positive favors home; negative favors away
Composite	-0.484	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.57-4.01	n/a	Base ML historical model prediction
variance_applied	5-4	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	0.515	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	5-3	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-3	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-3	n/a	Capped at max 12 realistic runs
simulation_result	5-3	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.57-4.01	5-4	0.441	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-4	5-4	-0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	5-4	5-3	1	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	5-3	5-3	-0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	5-3	5-3	-0	Capped at max 12 realistic runs
simulation_result	5-3	5-3	-0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
SF	Scott Bandura	batter / RF	better	predicted_AB 2.5
MIN	Ryan Kreidler	batter / SS	worse	predicted_H 2.41
MIN	Brooks Lee	batter / 3B	better	predicted_RBI 2.37
SF	Jonah Cox	batter / 2B	worse	predicted_AB 2.152
SF	Victor Bericoto	batter / 1B	worse	predicted_AB 2.121
MIN	Taylor Rogers	pitcher / Reliever	better	predicted_IP 2
SF	Bo Davidson	batter / DH	better	predicted_BB 2
SF	Joel Kuhnel	pitcher / Middle Reliever	better	predicted_IP 2

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	MIN pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	MIN
info	CURRENT_PITCHER_DUPLICATION	SF pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	SF
info	CURRENT_FINALIZED_MATCH_BOX	MIN model-prediction batting totals differ from final simulation box score (AB: current=34, sim=46; H: current=8, sim=10; 2B: current=1, sim=0; RBI: current=4, sim=5; R: current=4, sim=5; SO: current=7, sim=12).	MIN
info	CURRENT_PLAYER_MATCH_BOX	MIN current finalized hitter lines are model predictions and differ from the simulation box score: Luke Keaschall: AB current=4 box=5, H current=1 box=0, R current=1 box=0, SO current=0 box=2; Walker Jenkins: AB current=4 box=5, H current=1 box=0, R current=1 box=0, BB current=1 box=0; Brooks Lee: AB current=4 box=5, RBI current=1 box=3, BB current=0 box=1; Emmanuel Rodriguez: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Kody Clemens: AB current=4 box=6, H current=1 box=2, RBI current=1 box=2, SO current=1 box=2; Josh Bell: AB current=4 box=6, 2B current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Royce Lewis: R current=0 box=1, BB current=0 box=1, SO current=1 box=2; Victor Caratini: AB current=3 box=5, H current=0 box=2, R current=0 box=1	MIN
info	CURRENT_FINALIZED_MATCH_BOX	SF model-prediction batting totals differ from final simulation box score (AB: current=33, sim=37; H: current=7, sim=2; 2B: current=1, sim=0; RBI: current=4, sim=3; R: current=5, sim=3; BB: current=3, sim=5; SO: current=10, sim=9; SF: current=0, sim=1).	SF
info	CURRENT_PLAYER_MATCH_BOX	SF current finalized hitter lines are model predictions and differ from the simulation box score: Jonah Cox: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Victor Bericoto: AB current=4 box=5, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Turner Hill: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Bo Davidson: AB current=4 box=3, BB current=1 box=2; Scott Bandura: AB current=3 box=5; Shay Whitcomb: AB current=4 box=3, H current=1 box=0, RBI current=0 box=1, R current=0 box=1; Drew Gilbert: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Christian Koss: AB current=3 box=4, R current=1 box=0	SF
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

NYM@TEXMISS

Predicted: TEX 4-6 (conf +0.20, strong)Actual: NYM 6-3Brier 0.361total bias +1

Root cause: Backed the wrong side with real conviction

Evidence-backed diagnosis

published system edge 0.20 (strong); this is not the half-scale distance from 50%
team-run absolute error 5; total bias +1
pregame home-minus-away evidence: season +0.214, recent +0.404, pitcher -0.028, composite +0.517
confidence pipeline reached 0.601 before later publication/recency controls; published home probability was 0.601
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

TEX offense over-projected by 3
committed to TEX at 0.20 confidence and it lost

Team	Starter	Projected	Actual	ER err
NYM	Sean Manaea	2 ER / 17 outs	2 ER / 18 outs	0
TEX	MacKenzie Gore	2 ER / 14 outs	1 ER / 15 outs	-1

Decision internals: run factors home×0.99 / away×1.018; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Jake Burger (TEX 1B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.73]
Carson Benge (NYM RF): 2-for-4, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 1.42]
Wyatt Langford (TEX LF): 0-for-3, 0 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Corey Seager (TEX SS): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Juan Soto (NYM LF): 2-for-5, 0 HR, 1 RBI (projected 1.0 H, 1.0 HR) [run impact -0.79]

Pitcher versus batter: what actually broke this game

AWAY pitching

Sean Manaea (NYM) threw 18 outs, 2 ER = 3 ERA against a projected 3.18 ERA (-0.18 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The HOME lineup he faced produced 3 runs on 3 hits and 0 HR, against a projected 4 runs / 8 hits (-1 runs vs projection).

Who did the damage to him:

Danny Jansen: 0-for-2, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.263, weak - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

MacKenzie Gore (TEX) threw 15 outs, 1 ER = 1.8 ERA against a projected 3.86 ERA (-2.06 ERA error)

Bullpen behind him: 0 ER over 4 outs.

The AWAY lineup he faced produced 5 runs on 9 hits and 0 HR, against a projected 4 runs / 8 hits (1 runs vs projection).

Who did the damage to him:

Carson Benge: 2-for-4, 0 HR, 1 RBI [+1.42 runs vs projection] season xwOBA 0.339, above average - real projection error
Francisco Alvarez: 2-for-5, 0 HR, 2 RBI [+0.76 runs vs projection] season xwOBA 0.319, average - a good day, not a model error
A.J. Ewing: 1-for-3, 0 HR, 0 RBI [+0.33 runs vs projection] season xwOBA 0.32, average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
NYM	Sean Manaea (Starter)	18 / 17	3 / 3.18	2 / 2	3 / 3	0 / 0	2 / 4	1.05	9 / 6	4.23	venue away: 4.83 vs 4.04 ERA (worse here) · opp run factor 0.99 (average) · 31 prior appearances this season
NYM	Nathan Lavender (Setup Man)	3 / 3	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0.2	2 / 1	1.23
NYM	Dedniel Núñez (Middle Reliever (High Leverage))	3 / 1	0 / 54 (-54.00)	0 / 2	0 / 1	0 / 1	0 / 1	0	2 / 1	0.82
TEX	MacKenzie Gore (Starter)	15 / 14	1.8 / 3.86	1 / 2	4 / 3	0 / 1	1 / 1	2.91	2 / 3	3.8	venue home: 3.53 vs 5.79 ERA (ok) · opp run factor 1.02 (average) · 32 prior appearances this season
TEX	Jakob Junis (Setup Man)	4 / 4	0 / 13.5 (-13.50)	0 / 2	1 / 2	0 / 1	1 / 0	0.31	0 / 1	0.69

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Jake Burger (1B)	TEX	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	3 / 1	-1.73	0.32 average	no signal
Carson Benge (RF)	NYM	4 / 4	2 / 1	2 / 0	0 / 0	0 / 0	1 / 0	2 / 1	1 / 0	1 / 1	1.42	0.339 above average	projection too low
Wyatt Langford (LF)	TEX	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 1	3 / 1	-0.8	0.325 above average	bad day
Corey Seager (SS)	TEX	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	3 / 1	-0.8	0.352 elite	bad day
Juan Soto (LF)	NYM	5 / 4	2 / 1	0 / 0	0 / 0	0 / 1	1 / 1	0 / 1	0 / 1	0 / 1	-0.79	0.412 elite	bad day
Mark Vientos (1B)	NYM	2 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.78	0.312 average	no signal
Francisco Alvarez (DH)	NYM	5 / 4	2 / 1	2 / 0	0 / 0	0 / 0	2 / 0	2 / 0	0 / 1	1 / 1	0.76	0.319 average	no signal
Marcus Semien (2B)	NYM	5 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 0	-0.47	0.302 below average	—
Elias Díaz (DH)	TEX	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	0 / 1	-0.47	0.266 weak	—
Justin Foscue (2B)	TEX	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.328 above average	—
A.J. Ewing (CF)	NYM	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0.33	0.32 average	—
Bo Bichette (3B)	NYM	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	0.33	0.328 above average	—
Danny Jansen (C)	TEX	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 0	0 / 1	0.33	0.263 weak	—
Brandon Nimmo (RF)	TEX	4 / 4	2 / 1	0 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	0.16	0.342 elite	—
Ezequiel Duran (CF)	TEX	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	1 / 0	1 / 1	-0.14	0.303 average	—
Francisco Lindor (SS)	NYM	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0 / 1	0	0.33 above average	—
Luis Torrens (C)	NYM	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 0	0 / 1	0	0.294 below average	—
Josh Jung (3B)	TEX	4 / 4	1 / 1	1 / 1	0 / 0	0 / 0	1 / 0	0 / 1	0 / 0	2 / 0	0	0.347 elite	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.214	positive favors home; negative favors away
Recent form	0.404	positive favors home; negative favors away
Starting-pitcher matchup	-0.028	positive favors home; negative favors away
Composite	0.517	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.5-4.65	n/a	Base ML historical model prediction
variance_applied	5-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-6	0.601	Enhanced ML with recent form, H2H, low-score analysis
embedded_savant_run_environment	4-6	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-6	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-6	n/a	Capped at max 12 realistic runs
simulation_result	4-6	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.5-4.65	5-5	-0.151	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	5-5	4-6	2	Enhanced ML with recent form, H2H, low-score analysis
embedded_savant_run_environment	4-6	4-6	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-6	4-6	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-6	4-6	0	Capped at max 12 realistic runs
simulation_result	4-6	4-6	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
TEX	Jake Burger	batter / 1B	better	predicted_RBI 2.37
NYM	Marcus Semien	batter / 2B	worse	predicted_RBI 1.66
NYM	Francisco Lindor	batter / SS	better	predicted_RBI 1.52
TEX	Justin Foscue	batter / 2B	worse	predicted_AB 1.5
TEX	Corey Seager	batter / SS	better	predicted_RBI 1.46
NYM	Luis Torrens	batter / C	worse	predicted_SO 1.333
NYM	Juan Soto	batter / LF	worse	predicted_H -1.22
NYM	Bo Bichette	batter / 3B	worse	predicted_H -1.21

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	NYM pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	NYM
info	CURRENT_PITCHER_DUPLICATION	TEX pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	TEX
info	CURRENT_FINALIZED_MATCH_BOX	NYM model pitcher projections differ from final simulation pitching box score (H: current=7, sim=11; HR: current=2, sim=3).	NYM
info	CURRENT_FINALIZED_MATCH_BOX	NYM model-prediction batting totals differ from final simulation box score (AB: current=34, sim=32; H: current=8, sim=5; HR: current=1, sim=2; BB: current=3, sim=1; SO: current=8, sim=9).	NYM
info	CURRENT_PLAYER_MATCH_BOX	NYM current finalized hitter lines are model predictions and differ from the simulation box score: Mark Vientos: AB current=4 box=3, H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Carson Benge: SO current=1 box=0; Juan Soto: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Francisco Alvarez: AB current=4 box=3, H current=1 box=0, BB current=1 box=0; A.J. Ewing: H current=1 box=0, SO current=1 box=0; Francisco Lindor: H current=1 box=2, HR current=0 box=1, RBI current=1 box=2, BB current=1 box=0; Bo Bichette: H current=1 box=0, RBI current=1 box=0, SO current=1 box=2; Luis Torrens: H current=0 box=1, 2B current=0 box=1, SO current=1 box=2	NYM
info	CURRENT_FINALIZED_MATCH_BOX	TEX model-prediction batting totals differ from final simulation box score (AB: current=33, sim=35; H: current=8, sim=11; HR: current=1, sim=3; RBI: current=4, sim=6; R: current=4, sim=6; BB: current=3, sim=5; SO: current=8, sim=9).	TEX
info	CURRENT_PLAYER_MATCH_BOX	TEX current finalized hitter lines are model predictions and differ from the simulation box score: Wyatt Langford: H current=1 box=0, R current=1 box=0, SO current=1 box=2; Josh Jung: H current=1 box=2, RBI current=0 box=1, BB current=0 box=1, SO current=0 box=1; Ezequiel Duran: RBI current=1 box=0, R current=0 box=1; Corey Seager: AB current=4 box=5, H current=1 box=2, HR current=0 box=1, RBI current=1 box=2; Elias Díaz: H current=1 box=0, RBI current=1 box=0, BB current=0 box=1, SO current=1 box=0; Jake Burger: H current=1 box=2, HR current=1 box=2, RBI current=1 box=3, SO current=1 box=0; Justin Foscue: AB current=3 box=4, SO current=1 box=2; Danny Jansen: H current=0 box=1, 2B current=0 box=1, R current=0 box=1, BB current=0 box=1	TEX
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

SD@LADHIT

Predicted: LAD 4-6 (conf +0.06, lean)Actual: LAD 0-7Brier 0.221total bias +3

Root cause: Right winner, wrong score

Evidence-backed diagnosis

published system edge 0.06 (lean); this is not the half-scale distance from 50%
team-run absolute error 5; total bias +3
pregame home-minus-away evidence: season +0.593, recent -0.066, pitcher -0.041, composite +0.580
confidence pipeline reached 0.564 before later publication/recency controls; published home probability was 0.530
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 8 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 3 (10 vs 7)
SD offense over-projected by 4

Team	Starter	Projected	Actual	ER err
SD	Michael King	4 ER / 18 outs	5 ER / 12 outs	1
LAD	Brock Stewart	2 ER / 13 outs	0 ER / 3 outs	-2

Decision internals: run factors home×0.959 / away×1; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Tommy Edman (LAD 2B): 1-for-3, 1 HR, 2 RBI (projected 1.0 H, 0.0 HR) [run impact 1.26]
Max Muncy (LAD 3B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
Ty France (SD 1B): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.07]
Will Smith (LAD C): 1-for-4, 1 HR, 1 RBI (projected 1.0 H, 0.0 HR) [run impact 0.93]

Starter miss: Michael King (SD) posted 11.25 ERA vs projected 6.00 (5 ER in 12 outs); he is materially worse pitching away (3.31 ERA there vs 2.74 on the other side), a venue-role weakness the model under-weighted.

What to do: Enable/raise the venue-split weight (context.venue_adjustment_weight) and validate in shadow; the split is credible for this starter.

Pitcher versus batter: what actually broke this game

AWAY pitching

Michael King (SD) threw 12 outs, 5 ER = 11.25 ERA against a projected 6 ERA (5.25 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 6 runs on 7 hits and 2 HR, against a projected 4 runs / 7 hits (2 runs vs projection).

Who did the damage to him:

Tommy Edman: 1-for-3, 1 HR, 2 RBI [+1.26 runs vs projection] season xwOBA 0.304, average - a good day, not a model error
Will Smith: 1-for-4, 1 HR, 1 RBI [+0.93 runs vs projection] season xwOBA 0.385, elite - real projection error

Verdict: both: the starter was hit hard AND the lineup underproduced

HOME pitching

Brock Stewart (LAD) threw 3 outs, 0 ER = 0 ERA against a projected 4.15 ERA (-4.15 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The AWAY lineup he faced produced 0 runs on 4 hits and 0 HR, against a projected 6 runs / 7 hits (-6 runs vs projection).

No individual hitter beat his projection by a meaningful margin.

Verdict: neither side broke materially - the result is closer to variance

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
SD	Michael King (Starter)	12 / 18	11.25 / 6 (+5.25)	5 / 4	6 / 8	0 / 1	2 / 2	2.02	5 / 3	4.7	venue away: 3.31 vs 2.74 ERA (worse here) · opp run factor 0.96 (strong) · 31 prior appearances this season
LAD	Brock Stewart (Starter)	3 / 13	0 / 4.15 (-4.15)	0 / 2	1 / 3	0 / 1	0 / 4	1.44	1 / 5	3.18	opp run factor 1 (average) · 26 prior appearances this season
LAD	Jack Dreyer (Middle Reliever (High Leverage))	3 / 3	0 / 18 (-18.00)	0 / 2	0 / 1	0 / 1	0 / 1	0.41	1 / 0	0.83
LAD	Alex Vesia (Setup Man)	3 / 2	0 / 0	0 / 0	0 / 2	0 / 0	1 / 1	0.33	2 / 1	0.72

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Tommy Edman (2B)	LAD	3 / 4	1 / 1	0 / 0	0 / 0	1 / 0	2 / 0	2 / 0	1 / 0	1 / 1	1.26	0.304 average	no signal
Max Muncy (3B)	LAD	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-1.11	0.35 elite	bad day
Ty France (1B)	SD	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 0	2 / 1	-1.07	0.334 above average	bad day
Will Smith (C)	LAD	4 / 4	1 / 1	0 / 0	0 / 0	1 / 0	1 / 0	1 / 0	0 / 0	2 / 1	0.93	0.385 elite	projection too low
Dustin Harris (LF)	SD	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	-0.47	0.314 average	—
Jackson Merrill (CF)	SD	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	-0.47	0.348 elite	—
Manny Machado (3B)	SD	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	1 / 1	-0.47	0.33 above average	—
Austin Hays (DH)	SD	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.299 below average	—
Fernando Tatis Jr. (RF)	SD	4 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	2 / 1	-0.31	0.38 elite	—
Mookie Betts (SS)	LAD	4 / 4	1 / 1	1 / 0	0 / 0	0 / 1	1 / 1	1 / 1	1 / 0	0 / 0	-0.29	0.34 elite	—
Luis Campusano (C)	SD	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0.14	0.328 above average	—
Xander Bogaerts (SS)	SD	3 / 4	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0.14	0.323 above average	—
Freddie Freeman (1B)	LAD	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	3 / 0	0 / 0	0 / 1	2 / 1	0.14	0.366 elite	—
Kyle Tucker (RF)	LAD	4 / 3	1 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0.14	0.334 above average	—
Andy Pages (CF)	LAD	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 0	0 / 1	-0.14	0.34 elite	—
Jake Cronenworth (2B)	SD	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	0	0.275 weak	—
Teoscar Hernández (LF)	LAD	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 0	0 / 0	1 / 1	0	0.333 above average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.593	positive favors home; negative favors away
Recent form	-0.066	positive favors home; negative favors away
Starting-pitcher matchup	-0.041	positive favors home; negative favors away
Composite	0.58	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.75-5.28	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	0.53	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	4-6	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-6	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-6	n/a	Capped at max 12 realistic runs
simulation_result	4-6	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.75-5.28	4-5	-0.528	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	4-5	0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by matchup_recency_guardrail)
embedded_savant_run_environment	4-5	4-6	1	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-6	4-6	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-6	4-6	0	Capped at max 12 realistic runs
simulation_result	4-6	4-6	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
SD	Luis Campusano	batter / C	worse	predicted_RBI 2.56
LAD	Max Muncy	batter / 3B	better	predicted_RBI 2.29
SD	Austin Hays	batter / DH	worse	predicted_BB 1.968
SD	Dustin Harris	batter / LF	worse	predicted_BB 1.765
LAD	Kyle Tucker	batter / RF	better	predicted_H 1.66
LAD	Will Smith	batter / C	worse	predicted_AB 1.636
LAD	Teoscar Hernández	batter / LF	deboosted	predicted_RBI 1.53
SD	Jackson Merrill	batter / CF	worse	predicted_H -1.36

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	LAD pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	LAD
info	CURRENT_PITCHER_DUPLICATION	SD pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	SD
info	CURRENT_FINALIZED_MATCH_BOX	LAD model-prediction batting totals differ from final simulation box score (AB: current=34, sim=36; H: current=7, sim=11; 2B: current=1, sim=4; 3B: current=0, sim=1; RBI: current=4, sim=6; R: current=5, sim=6; SO: current=7, sim=3).	LAD
info	CURRENT_PLAYER_MATCH_BOX	LAD current finalized hitter lines are model predictions and differ from the simulation box score: Will Smith: AB current=4 box=5, H current=1 box=0, SO current=1 box=0; Freddie Freeman: H current=1 box=2, 2B current=0 box=1, R current=0 box=1, SO current=1 box=0; Max Muncy: H current=1 box=3, 2B current=1 box=2, HR current=0 box=1, RBI current=1 box=3; Josue De Paula: AB current=3 box=4, R current=1 box=0, BB current=1 box=0, SO current=0 box=1; Mookie Betts: H current=1 box=2, HR current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Kyle Tucker: AB current=3 box=4, H current=0 box=2, 2B current=0 box=1, RBI current=0 box=1; Teoscar Hernández: H current=1 box=2, 3B current=0 box=1, RBI current=1 box=2, R current=0 box=1; Andy Pages: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, BB current=0 box=1	LAD
info	CURRENT_FINALIZED_MATCH_BOX	SD model pitcher projections differ from final simulation pitching box score (H: current=9, sim=11; BB: current=3, sim=4).	SD
info	CURRENT_FINALIZED_MATCH_BOX	SD model-prediction batting totals differ from final simulation box score (AB: current=33, sim=34; 2B: current=1, sim=2; HR: current=1, sim=2; R: current=6, sim=4; BB: current=3, sim=8; SO: current=9, sim=8; HBP: current=0, sim=1).	SD
info	CURRENT_PLAYER_MATCH_BOX	SD current finalized hitter lines are model predictions and differ from the simulation box score: Fernando Tatis Jr.: 2B current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Ty France: HR current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=0 box=1; Dustin Harris: AB current=4 box=3, H current=1 box=0, R current=1 box=0, BB current=0 box=2; Jackson Merrill: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Manny Machado: H current=1 box=2, HR current=0 box=1, R current=0 box=1, SO current=1 box=0; Austin Hays: 2B current=0 box=1, R current=0 box=1, BB current=0 box=2; Luis Campusano: AB current=3 box=4, HR current=0 box=1, RBI current=0 box=3, R current=0 box=1; Xander Bogaerts: AB current=4 box=3, H current=0 box=1, 2B current=0 box=1, R current=1 box=0	SD
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

STL@PITHIT

Predicted: PIT 4-5 (conf +0.08, lean)Actual: PIT 0-2Brier 0.213total bias +7

Root cause: Right winner, wrong score

Evidence-backed diagnosis

published system edge 0.08 (lean); this is not the half-scale distance from 50%
team-run absolute error 7; total bias +7
pregame home-minus-away evidence: season +0.333, recent +0.490, pitcher -0.026, composite +0.428
confidence pipeline reached 0.553 before later publication/recency controls; published home probability was 0.539
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 7 (9 vs 2)
STL offense over-projected by 4
PIT offense over-projected by 3

Team	Starter	Projected	Actual	ER err
STL	Andre Pallante	2 ER / 16 outs	1 ER / 18 outs	-1
PIT	Jared Jones	1 ER / 15 outs	0 ER / 21 outs	-1

Decision internals: run factors home×0.981 / away×1.007; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Jordan Walker (STL RF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Brandon Lowe (PIT 2B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
Alec Burleson (STL 1B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -1.11]
JJ Wetherholt (STL 2B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Iván Herrera (STL DH): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]

Pitcher versus batter: what actually broke this game

AWAY pitching

Andre Pallante (STL) threw 18 outs, 1 ER = 1.5 ERA against a projected 3.38 ERA (-1.88 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 2 runs on 5 hits and 0 HR, against a projected 5 runs / 9 hits (-3 runs vs projection).

Who did the damage to him:

Konnor Griffin: 2-for-4, 0 HR, 0 RBI [+0.78 runs vs projection] season xwOBA 0.307, average - a good day, not a model error
Spencer Horwitz: 1-for-2, 0 HR, 0 RBI [+0.31 runs vs projection] season xwOBA 0.313, average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

HOME pitching

Jared Jones (PIT) threw 21 outs, 0 ER = 0 ERA against a projected 1.8 ERA (-1.8 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The AWAY lineup he faced produced 0 runs on 1 hits and 0 HR, against a projected 4 runs / 8 hits (-4 runs vs projection).

Who did the damage to him:

Nathan Church: 1-for-3, 0 HR, 0 RBI [+0.31 runs vs projection] season xwOBA 0.288, below average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
STL	Andre Pallante (Starter)	18 / 16	1.5 / 3.38	1 / 2	3 / 2	0 / 2	2 / 1	1.83	2 / 10	2.86	venue away: 2.75 vs 4.41 ERA (ok) · opp run factor 0.98 (average) · 27 prior appearances this season
PIT	Jared Jones (Starter)	21 / 15	0 / 1.8	0 / 1	1 / 3	0 / 0	0 / 0	1.91	9 / 4	6.94	venue home: 4.68 vs 3.54 ERA (worse here) · opp run factor 1.01 (average) · 19 prior appearances this season
PIT	Gregory Soto (Middle Reliever (High Leverage))	3 / 3	0 / 9 (-9.00)	0 / 1	0 / 2	0 / 0	1 / 0	0.55	2 / 1	1.36
PIT	Luke Weaver (Setup Man)	3 / 3	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	0.22	3 / 1	1.06

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Jordan Walker (RF)	STL	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	0 / 1	-1.4	0.342 elite	bad day
Brandon Lowe (2B)	PIT	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	0 / 1	-1.4	0.328 above average	bad day
Alec Burleson (1B)	STL	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	0 / 1	-1.11	0.373 elite	bad day
JJ Wetherholt (2B)	STL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	3 / 1	-0.8	0.349 elite	bad day
Iván Herrera (DH)	STL	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	4 / 1	-0.8	0.335 above average	bad day
Leonardo Bernal (C)	STL	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	2 / 1	-0.78	0.379 elite	bad day
Konnor Griffin (SS)	PIT	4 / 4	2 / 1	0 / 1	1 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	0.78	0.307 average	no signal
Oneil Cruz (CF)	PIT	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	1 / 1	1 / 1	1 / 1	-0.47	0.353 elite	—
Jacob Gonzalez (3B)	PIT	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	-0.47	0.305 average	—
Ronny Simon (RF)	PIT	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	-0.47	0.328 above average	—
Bryan Reynolds (LF)	PIT	3 / 4	1 / 1	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	-0.33	0.354 elite	—
Nathan Church (CF)	STL	3 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	0.31	0.288 below average	—
Spencer Horwitz (1B)	PIT	2 / 4	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	0 / 0	0.31	0.313 average	—
Thomas Saggese (3B)	STL	2 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	-0.14	0.314 average	—
Masyn Winn (SS)	STL	2 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	1 / 1	-0.14	0.297 below average	—
Rafael Flores Jr. (C)	PIT	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	1 / 1	-0.14	0.362 elite	—
Bryan Torres (LF)	STL	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0	0.289 below average	—
Ryan O'Hearn (DH)	PIT	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	0 / 0	1 / 1	0	0.316 average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.333	positive favors home; negative favors away
Recent form	0.49	positive favors home; negative favors away
Starting-pitcher matchup	-0.026	positive favors home; negative favors away
Composite	0.428	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.46-5.4	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	0.539	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-5	n/a	Capped at max 12 realistic runs
simulation_result	4-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.46-5.4	4-5	0.056	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	4-5	0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-5	4-5	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-5	4-5	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-5	4-5	0	Capped at max 12 realistic runs
simulation_result	4-5	4-5	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
STL	Bryan Torres	batter / LF	worse	predicted_H 2.4
STL	Masyn Winn	batter / SS	worse	predicted_SO 2.16
PIT	Brandon Lowe	batter / 2B	worse	predicted_RBI 2.08
PIT	Oneil Cruz	batter / CF	worse	predicted_AB -1.79
STL	Jordan Walker	batter / RF	deboosted	predicted_RBI 1.54
PIT	Ronny Simon	batter / RF	worse	predicted_SO 1.533
STL	Riley O'Brien	pitcher / Closer	better	predicted_IP 1.433
PIT	Spencer Horwitz	batter / 1B	better	predicted_AB -1.404

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	PIT pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	PIT
info	CURRENT_PITCHER_DUPLICATION	STL pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	STL
info	CURRENT_FINALIZED_MATCH_BOX	PIT model-prediction batting totals differ from final simulation box score (AB: current=35, sim=28; H: current=9, sim=4; 2B: current=2, sim=0; HR: current=1, sim=3; BB: current=3, sim=2; SO: current=8, sim=12).	PIT
info	CURRENT_PLAYER_MATCH_BOX	PIT current finalized hitter lines are model predictions and differ from the simulation box score: Konnor Griffin: 2B current=1 box=0, RBI current=1 box=0, SO current=1 box=2; Oneil Cruz: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Ryan O'Hearn: AB current=4 box=3, HR current=0 box=1, R current=0 box=1, SO current=1 box=0; Brandon Lowe: RBI current=1 box=3, SO current=1 box=2; Bryan Reynolds: H current=1 box=0, 2B current=1 box=0, R current=1 box=0, BB current=1 box=0; Jacob Gonzalez: H current=1 box=0, SO current=1 box=0; Rafael Flores Jr.: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0; Ronny Simon: AB current=4 box=2, H current=1 box=0, BB current=0 box=1, SO current=1 box=2	PIT
info	CURRENT_FINALIZED_MATCH_BOX	STL model-prediction batting totals differ from final simulation box score (AB: current=33, sim=34; 3B: current=0, sim=1; RBI: current=5, sim=4; BB: current=3, sim=1; SF: current=0, sim=1).	STL
info	CURRENT_PLAYER_MATCH_BOX	STL current finalized hitter lines are model predictions and differ from the simulation box score: JJ Wetherholt: R current=1 box=0, BB current=1 box=0, SO current=1 box=0; Iván Herrera: H current=1 box=2, 2B current=0 box=1, RBI current=1 box=0, R current=1 box=0; Bryan Torres: AB current=3 box=4, H current=0 box=3, 3B current=0 box=1, R current=0 box=1; Thomas Saggese: AB current=4 box=2, H current=1 box=0, RBI current=0 box=1, R current=0 box=1; Jordan Walker: RBI current=1 box=2; Leonardo Bernal: SO current=1 box=0; Alec Burleson: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, BB current=1 box=0; Nathan Church: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0	STL
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

TB@NYYHIT

Predicted: NYY 3-4 (conf +0.07, lean)Actual: NYY 0-2Brier 0.216total bias +5

Root cause: Right winner, wrong score

Evidence-backed diagnosis

published system edge 0.07 (lean); this is not the half-scale distance from 50%
team-run absolute error 5; total bias +5
pregame home-minus-away evidence: season +0.082, recent -0.234, pitcher -0.080, composite -0.009
confidence pipeline reached 0.548 before later publication/recency controls; published home probability was 0.535
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 5 (7 vs 2)
TB offense over-projected by 3
Carlos Rodón (NYY) allowed 3 more ER than projected (4 vs 1)

Team	Starter	Projected	Actual	ER err
TB	Nick Martinez	1 ER / 18 outs	1 ER / 20 outs	0
NYY	Carlos Rodón	1 ER / 17 outs	4 ER / 20 outs	3

Decision internals: run factors home×0.966 / away×1.046; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Junior Caminero (TB 3B): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.73]
Yandy Díaz (TB DH): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Liam Hicks (TB C): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Anthony Volpe (NYY 2B): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Jonny DeLuca (TB CF): 0-for-3, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.78]

Pitcher versus batter: what actually broke this game

AWAY pitching

Nick Martinez (TB) threw 18 outs, 2 ER = 3 ERA against a projected 1.5 ERA (1.5 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The HOME lineup he faced produced 2 runs on 5 hits and 1 HR, against a projected 4 runs / 8 hits (-2 runs vs projection).

No individual hitter beat his projection by a meaningful margin.

Verdict: offense: the lineup scored well below its projection

HOME pitching

Carlos Rodón (NYY) threw 24 outs, 0 ER = 0 ERA against a projected 1.59 ERA (-1.59 ERA error)

Bullpen behind him: 0 ER over 3 outs.

The AWAY lineup he faced produced 0 runs on 3 hits and 0 HR, against a projected 5 runs / 7 hits (-5 runs vs projection).

No individual hitter beat his projection by a meaningful margin.

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
TB	Nick Martinez (Starter)	18 / 18	3 / 1.5	2 / 1	5 / 5	1 / 0	1 / 0	0.53	7 / 5	2.96	venue away: 1.99 vs 3.9 ERA (ok) · opp run factor 0.97 (strong) · 29 prior appearances this season
NYY	Carlos Rodón (Starter)	24 / 17	0 / 1.59	0 / 1	3 / 1	0 / 1	0 / 1	2.02	8 / 7	5.62	venue home: 2.61 vs 3.18 ERA (ok) · opp run factor 1.05 (strong) · 15 prior appearances this season
NYY	David Bednar (Closer)	3 / 3	0 / 9 (-9.00)	0 / 1	0 / 2	0 / 0	0 / 1	0.19	0 / 1	1.16

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Junior Caminero (3B)	TB	3 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	1 / 1	-1.73	0.363 elite	bad day
Yandy Díaz (DH)	TB	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	2 / 1	-0.8	0.347 elite	bad day
Liam Hicks (C)	TB	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	1 / 1	-0.8	0.32 average	no signal
Anthony Volpe (2B)	NYY	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	3 / 1	-0.8	0.296 below average	projection too high
Jonny DeLuca (CF)	TB	3 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.78	0.259 weak	projection too high
Ben Rice (DH)	NYY	3 / 4	1 / 1	1 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 1	1 / 1	-0.62	0.376 elite	bad day
Heliot Ramos (RF)	NYY	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	2 / 1	-0.47	0.329 above average	—
Luis García Jr. (1B)	NYY	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.47	0.345 elite	—
Jonathan Aranda (1B)	TB	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.33	0.345 elite	—
Cody Bellinger (LF)	NYY	4 / 4	1 / 1	0 / 1	0 / 0	1 / 0	2 / 1	1 / 1	0 / 1	0 / 0	0.29	0.338 above average	—
Chandler Simpson (LF)	TB	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0	0.266 weak	—
Richie Palacios (2B)	TB	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	0	0.31 average	—
Jorge Mateo (SS)	TB	2 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.28 weak	—
Victor Mesa Jr. (RF)	TB	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	0	0.333 above average	—
George Lombard Jr. (SS)	NYY	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 1	0	0.298 below average	—
José Caballero (3B)	NYY	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	2 / 1	0	0.282 weak	—
Ali Sánchez (C)	NYY	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.252 weak	—
Spencer Jones (CF)	NYY	3 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	0 / 1	0	0.328 above average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.082	positive favors home; negative favors away
Recent form	-0.234	positive favors home; negative favors away
Starting-pitcher matchup	-0.08	positive favors home; negative favors away
Composite	-0.009	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.81-5.21	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	3-4	0.535	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	3-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-4	n/a	Capped at max 12 realistic runs
simulation_result	3-4	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.81-5.21	4-5	-0.398	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	3-4	0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	3-4	3-4	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-4	3-4	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-4	3-4	0	Capped at max 12 realistic runs
simulation_result	3-4	3-4	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
TB	Yandy Díaz	batter / DH	worse	predicted_SO 2.358
TB	Alex Cook	pitcher / Reliever	worse	predicted_ER 2
TB	Junior Caminero	batter / 3B	worse	predicted_AB -1.63
TB	Chandler Simpson	batter / LF	worse	predicted_H -1.48
NYY	Cody Bellinger	batter / LF	worse	predicted_H -1.46
NYY	Luis García Jr.	batter / 1B	better	predicted_RBI 1.45
TB	Jorge Mateo	batter / SS	worse	predicted_AB 1.356
NYY	Anthony Volpe	batter / 2B	worse	predicted_SO 1.222

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	NYY pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	NYY
info	CURRENT_PITCHER_DUPLICATION	TB pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	TB
info	CURRENT_FINALIZED_MATCH_BOX	NYY model-prediction batting totals differ from final simulation box score (AB: current=32, sim=37; H: current=8, sim=7; 2B: current=1, sim=0; HR: current=1, sim=2; BB: current=3, sim=2; HBP: current=0, sim=2).	NYY
info	CURRENT_PLAYER_MATCH_BOX	NYY current finalized hitter lines are model predictions and differ from the simulation box score: Heliot Ramos: AB current=3 box=4, R current=1 box=0, HBP current=0 box=1; George Lombard Jr.: AB current=4 box=5, H current=1 box=2, RBI current=0 box=1, SO current=1 box=2; Ben Rice: AB current=4 box=5, BB current=1 box=0, SO current=1 box=0; Anthony Volpe: H current=1 box=0, SO current=1 box=2; Cody Bellinger: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, R current=1 box=0; José Caballero: AB current=3 box=4, H current=1 box=0; Luis García Jr.: AB current=4 box=5, H current=1 box=2, HR current=0 box=1, RBI current=1 box=2; Ali Sánchez: H current=0 box=1, R current=0 box=1, BB current=0 box=1	NYY
info	CURRENT_FINALIZED_MATCH_BOX	TB model-prediction batting totals differ from final simulation box score (H: current=7, sim=4; RBI: current=5, sim=3; R: current=5, sim=3; BB: current=4, sim=3; SO: current=8, sim=10).	TB
info	CURRENT_PLAYER_MATCH_BOX	TB current finalized hitter lines are model predictions and differ from the simulation box score: Jonathan Aranda: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=1 box=0; Yandy Díaz: AB current=4 box=5, H current=1 box=0, RBI current=1 box=0, R current=1 box=0; Junior Caminero: AB current=4 box=3, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Jonny DeLuca: 2B current=1 box=0, R current=0 box=1, SO current=1 box=0; Liam Hicks: AB current=4 box=3, H current=1 box=0, RBI current=1 box=0; Chandler Simpson: H current=1 box=0, SO current=0 box=1; Richie Palacios: H current=0 box=1, 2B current=0 box=1, RBI current=0 box=1, BB current=0 box=1; Jorge Mateo: AB current=3 box=4, H current=0 box=1, HR current=0 box=1, RBI current=0 box=1	TB
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

TB@NYYMISS

Predicted: NYY 4-5 (conf +0.04, lean)Actual: TB 6-1Brier 0.271total bias +2

Root cause: Small edge on the losing side

Evidence-backed diagnosis

published system edge 0.04 (lean); this is not the half-scale distance from 50%
team-run absolute error 6; total bias +2
pregame home-minus-away evidence: season +0.082, recent -0.234, pitcher -0.010, composite +0.145
confidence pipeline reached 0.529 before later publication/recency controls; published home probability was 0.521
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

NYY offense over-projected by 4

Decision internals: run factors home×0.986 / away×0.996; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

Ben Rice (NYY DH): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.73]
Junior Caminero (TB 3B): 1-for-5, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -0.95]
Jonathan Aranda (TB 1B): 0-for-1, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.8]
Luis García Jr. (NYY 1B): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 0.0 HR) [run impact -0.78]

Pitcher versus batter: what actually broke this game

AWAY pitching

Drew Rasmussen (TB) threw 20 outs, 1 ER = 1.35 ERA against a projected 3 ERA (-1.65 ERA error)

Bullpen behind him: 0 ER over 6 outs.

The HOME lineup he faced produced 1 runs on 2 hits and 0 HR, against a projected 4 runs / 6 hits (-3 runs vs projection).

Who did the damage to him:

Heliot Ramos: 1-for-4, 0 HR, 0 RBI [+0.31 runs vs projection] season xwOBA 0.329, above average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

HOME pitching

Max Fried (NYY) threw 20 outs, 4 ER = 5.4 ERA against a projected 4.5 ERA (0.9 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The AWAY lineup he faced produced 1 runs on 5 hits and 0 HR, against a projected 3 runs / 5 hits (-2 runs vs projection).

No individual hitter beat his projection by a meaningful margin.

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
TB	Drew Rasmussen (Starter)	20 / 18	1.35 / 3	1 / 2	1 / 5	0 / 1	3 / 0	0.82	11 / 5	4.31	venue away: 2.57 vs 2.87 ERA (ok) · opp run factor 0.99 (strong) · 29 prior appearances this season
TB	Bryan Baker (Closer)	3 / 3	0 / 18 (-18.00)	0 / 2	1 / 3	0 / 1	1 / 0	0.27	1 / 0	0.72
TB	Cam Booser (Middle Reliever (High Leverage))	3 / 3	0 / 9 (-9.00)	0 / 1	0 / 2	0 / 0	0 / 0	0.12	1 / 0	0.86
NYY	Max Fried (Starter)	20 / 18	5.4 / 4.5	4 / 3	6 / 6	0 / 0	2 / 5	1.32	4 / 5	5.57	venue home: 3.66 vs 2.34 ERA (worse here) · opp run factor 1 (strong) · 19 prior appearances this season

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
Ben Rice (DH)	NYY	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	3 / 1	-1.73	0.376 elite	bad day
Junior Caminero (3B)	TB	5 / 4	1 / 1	1 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	0 / 1	-0.95	0.363 elite	bad day
Jonathan Aranda (1B)	TB	1 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	1 / 1	-0.8	0.345 elite	bad day
Luis García Jr. (1B)	NYY	4 / 4	0 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	1 / 1	-0.78	0.345 elite	bad day
Cody Bellinger (LF)	NYY	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	1 / 1	-0.47	0.338 above average	—
Heliot Ramos (RF)	NYY	4 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0 / 0	2 / 1	0.31	0.329 above average	—
Yandy Díaz (DH)	TB	5 / 4	2 / 1	0 / 0	0 / 0	0 / 0	4 / 1	0 / 1	0 / 1	0 / 1	0.14	0.347 elite	—
Spencer Jones (CF)	NYY	3 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	1 / 0	2 / 2	-0.14	0.328 above average	—
Jonny DeLuca (CF)	TB	4 / 4	1 / 1	1 / 1	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	3 / 1	0	0.259 weak	—
Chandler Simpson (LF)	TB	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 0	0	0.266 weak	—
José Caballero (3B)	NYY	0 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0	0.282 weak	—
George Lombard Jr. (SS)	NYY	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	1 / 0	0 / 1	0 / 0	1 / 1	0	0.298 below average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.082	positive favors home; negative favors away
Recent form	-0.234	positive favors home; negative favors away
Starting-pitcher matchup	-0.01	positive favors home; negative favors away
Composite	0.145	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	4.35-4.77	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	0.521	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-5	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-5	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-5	n/a	Capped at max 12 realistic runs
simulation_result	4-5	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	4.35-4.77	4-5	0.583	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	4-5	0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run, matchup_recency_guardrail)
embedded_savant_run_environment	4-5	4-5	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	4-5	4-5	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	4-5	4-5	0	Capped at max 12 realistic runs
simulation_result	4-5	4-5	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
NYY	José Caballero	batter / 3B	better	predicted_RBI 2.583
TB	Jonathan Aranda	batter / 1B	neutral	predicted_RBI 2.2
NYY	Spencer Jones	batter / CF	better	predicted_H 2.09
TB	Yandy Díaz	batter / DH	worse	predicted_BB 1.628
NYY	Ali Sánchez	batter / C	worse	predicted_AB 1.5
TB	Jorge Mateo	batter / SS	worse	predicted_AB 1.356
NYY	Cody Bellinger	batter / LF	worse	predicted_AB -1.17
TB	Liam Hicks	batter / C	worse	predicted_H -1.14

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	NYY pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	NYY
info	CURRENT_PITCHER_DUPLICATION	TB pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	TB
info	CURRENT_FINALIZED_MATCH_BOX	NYY model-prediction batting totals differ from final simulation box score (AB: current=32, sim=36; H: current=7, sim=10; 2B: current=1, sim=4; HR: current=1, sim=2; RBI: current=4, sim=5; R: current=4, sim=5; BB: current=3, sim=0; SO: current=10, sim=5).	NYY
info	CURRENT_PLAYER_MATCH_BOX	NYY current finalized hitter lines are model predictions and differ from the simulation box score: Heliot Ramos: AB current=3 box=4, H current=1 box=0, R current=1 box=0, SO current=1 box=0; Luis García Jr.: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0; Ben Rice: HR current=1 box=0, RBI current=1 box=0, BB current=1 box=0, SO current=1 box=2; Cody Bellinger: 2B current=0 box=1, RBI current=1 box=0, BB current=1 box=0, SO current=1 box=0; Anthony Volpe: AB current=3 box=4, 2B current=0 box=1, BB current=1 box=0; José Caballero: AB current=3 box=4, H current=0 box=3, 2B current=0 box=1, HR current=0 box=1; Ali Sánchez: AB current=3 box=4, SO current=1 box=0; Spencer Jones: H current=1 box=3, 2B current=0 box=1, HR current=0 box=1, RBI current=1 box=2	NYY
info	CURRENT_FINALIZED_MATCH_BOX	TB model-prediction batting totals differ from final simulation box score (H: current=8, sim=7; HR: current=1, sim=0; RBI: current=5, sim=4; R: current=5, sim=4; BB: current=3, sim=5; SO: current=8, sim=7; HBP: current=0, sim=1).	TB
info	CURRENT_PLAYER_MATCH_BOX	TB current finalized hitter lines are model predictions and differ from the simulation box score: Jonathan Aranda: 2B current=0 box=1, RBI current=1 box=3; Liam Hicks: H current=1 box=0, RBI current=1 box=0, BB current=0 box=1; Jonny DeLuca: H current=1 box=0, 2B current=1 box=0, RBI current=1 box=0, SO current=1 box=0; Junior Caminero: AB current=4 box=5, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Yandy Díaz: AB current=4 box=3, H current=1 box=0, BB current=1 box=2, SO current=1 box=2; Victor Mesa Jr.: AB current=4 box=3, H current=1 box=2, HBP current=0 box=1; Richie Palacios: BB current=0 box=1, SO current=1 box=0; Jorge Mateo: AB current=3 box=4, H current=0 box=1	TB
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

WAS@DETMISS

Predicted: DET 3-4 (conf +0.14, moderate)Actual: WAS 3-1Brier 0.326total bias +3

Root cause: Backed the wrong side with real conviction

Evidence-backed diagnosis

published system edge 0.14 (moderate); this is not the half-scale distance from 50%
team-run absolute error 3; total bias +3
pregame home-minus-away evidence: season +0.390, recent +0.618, pitcher +0.017, composite +0.587
confidence pipeline reached 0.593 before later publication/recency controls; published home probability was 0.571
score-volatility actions: no_change
forecast-file integrity: clean (0 defects, 7 informational note(s)); the saved artifact is not why the pick missed

Quantified contributing factors

total runs over-projected by 3 (7 vs 4)
DET offense over-projected by 3
committed to DET at 0.14 confidence and it lost

Team	Starter	Projected	Actual	ER err
WAS	Riley Cornelio	2 ER / 12 outs	0 ER / 8 outs	-2
DET	Drew Anderson	3 ER / 15 outs	3 ER / 18 outs	0

Decision internals: run factors home×0.994 / away×0.977; savant lineup adjustment applied

Player-level breakdown (projected vs actual)

Every value is actual / projected. Run impact approximates how far each hitter's real line moved run expectancy from its projection; it orders the movers and is not a published stat.

Top deviations that drove the game:

CJ Abrams (WAS 2B): 2-for-4, 2 HR, 3 RBI (projected 1.0 H, 0.0 HR) [run impact 1.69]
Eduardo Valencia (DET C): 0-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.4]
James Wood (WAS RF): 1-for-4, 0 HR, 0 RBI (projected 1.0 H, 1.0 HR) [run impact -1.26]
Spencer Torkelson (DET 1B): 0-for-0, 0 HR, 0 RBI (projected 0.0 H, 0.0 HR) [run impact 0.66]

Starter miss: Riley Cornelio (WAS) posted 0.00 ERA vs projected 4.50 (0 ER in 8 outs); no credible venue, opponent-strength or experience signal explains it — an outlier start (variance).

What to do: Variance — no model change is warranted for this start.

Pitcher versus batter: what actually broke this game

AWAY pitching

Riley Cornelio (WAS) threw 8 outs, 0 ER = 0 ERA against a projected 4.5 ERA (-4.5 ERA error)

Bullpen behind him: 0 ER over 7 outs.

The HOME lineup he faced produced 1 runs on 7 hits and 0 HR, against a projected 4 runs / 7 hits (-3 runs vs projection).

Who did the damage to him:

Spencer Torkelson: 0-for-0, 0 HR, 0 RBI [+0.66 runs vs projection] season xwOBA 0.312, average - a good day, not a model error
Kevin McGonigle: 2-for-5, 0 HR, 0 RBI [+0.45 runs vs projection] season xwOBA 0.359, elite - a good day, not a model error
Hao-Yu Lee: 1-for-3, 0 HR, 0 RBI [+0.35 runs vs projection] season xwOBA 0.301, below average - a good day, not a model error

Verdict: neither side broke materially - the result is closer to variance

HOME pitching

Drew Anderson (DET) threw 18 outs, 3 ER = 4.5 ERA against a projected 5.4 ERA (-0.9 ERA error)

Bullpen behind him: 0 ER over 0 outs.

The AWAY lineup he faced produced 3 runs on 7 hits and 2 HR, against a projected 4 runs / 7 hits (-1 runs vs projection).

Who did the damage to him:

CJ Abrams: 2-for-4, 2 HR, 3 RBI [+1.69 runs vs projection] season xwOBA 0.327, above average - real projection error
Dylan Crews: 2-for-4, 0 HR, 0 RBI [+0.47 runs vs projection] season xwOBA 0.317, average - a good day, not a model error

Verdict: offense: the lineup scored well below its projection

Pitcher	Team	Outs	ERA	ER	H	HR	BB	exp BB	SO	exp SO	Context (as-of)
WAS	Riley Cornelio (Starter)	8 / 12	0 / 4.5 (-4.50)	0 / 2	1 / 5	0 / 1	3 / 2	0.97	4 / 7	7.26	venue away: 0 vs 0 ERA (ok) · opp run factor 0.99 (average) · 16 prior appearances this season
WAS	Erik Tolman (Closer)	3 / 4	0 / 6.75 (-6.75)	0 / 1	0 / 2	0 / 0	1 / 0	0.44	3 / 2	0.91
WAS	Jack Sinclair (Reliever)	4 / 0	0 / n/a	0 / 0	0 / 1	0 / 0	0 / 0	n/a	1 / 1	n/a
DET	Drew Anderson (Starter)	18 / 15	4.5 / 5.4	3 / 3	5 / 3	2 / 1	0 / 3	2.82	5 / 9	4.44	venue home: 2.66 vs 2.04 ERA (worse here) · opp run factor 0.98 (strong) · 48 prior appearances this season

Hitters: actual / projected (sorted by run impact vs projection)

Batter	Team	AB	H	2B	3B	HR	RBI	R	BB	SO	Run impact	Season xwOBA	Signal
CJ Abrams (2B)	WAS	4 / 4	2 / 1	0 / 1	0 / 0	2 / 0	3 / 1	2 / 1	0 / 1	1 / 1	1.69	0.327 above average	projection too low
Eduardo Valencia (C)	DET	4 / 4	0 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 0	1 / 1	-1.4	0.351 elite	bad day
James Wood (RF)	WAS	4 / 4	1 / 1	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	0 / 1	2 / 1	-1.26	0.408 elite	bad day
Spencer Torkelson (1B)	DET	0 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	3 / 1	0 / 1	0.66	0.312 average	no signal
Dylan Crews (CF)	WAS	4 / 4	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	1 / 1	0.47	0.317 average	—
Yohandy Morales (DH)	WAS	3 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 1	-0.47	0.336 above average	—
Daylen Lile (LF)	WAS	4 / 4	0 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 0	1 / 1	-0.47	0.327 above average	—
Javier Báez (LF)	DET	4 / 3	0 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	-0.47	0.263 weak	—
Kevin McGonigle (3B)	DET	5 / 4	2 / 1	1 / 0	0 / 0	0 / 0	0 / 1	0 / 1	0 / 1	3 / 0	0.45	0.359 elite	—
Hao-Yu Lee (2B)	DET	3 / 4	1 / 1	0 / 1	0 / 0	0 / 0	0 / 1	0 / 0	2 / 0	1 / 1	0.35	0.301 below average	—
Jorbit Vivas (3B)	WAS	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	-0.33	0.294 below average	—
Ben Malgeri (RF)	DET	3 / 3	1 / 1	1 / 0	0 / 0	0 / 0	0 / 0	1 / 0	0 / 0	0 / 1	0.31	0.318 average	—
Max Clark (CF)	DET	4 / 3	2 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	0.14	0.321 above average	—
Nasim Nuñez (SS)	WAS	3 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.267 weak	—
Andrés Chaparro (1B)	WAS	4 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 0	0 / 0	2 / 1	0	0.345 elite	—
Keibert Ruiz (C)	WAS	3 / 3	1 / 1	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0	0.288 below average	—
Gleyber Torres (DH)	DET	4 / 4	1 / 1	0 / 0	0 / 0	0 / 0	0 / 1	0 / 1	1 / 1	2 / 1	0	0.324 above average	—
John Peck (SS)	DET	4 / 3	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	0 / 0	1 / 1	0	0.33 above average	—

Full model trace, player evidence, and audit findings

Fundamental evidence entering the pick

Signal	Home-away edge	Meaning
Season quality	0.39	positive favors home; negative favors away
Recent form	0.618	positive favors home; negative favors away
Starting-pitcher matchup	0.017	positive favors home; negative favors away
Composite	0.587	positive favors home; negative favors away

Score and confidence evolution

Stage	Away-home runs	Confidence	What changed
historical_ml	3.54-4.5	n/a	Base ML historical model prediction
variance_applied	4-5	n/a	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	3-4	0.571	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run)
embedded_savant_run_environment	3-4	n/a	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-4	n/a	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-4	n/a	Capped at max 12 realistic runs
simulation_result	3-4	n/a	Final inning-by-inning simulation result

Official postgame stage attribution

Stage	Before	After	Toward final pick	What changed
variance_applied	3.54-4.5	4-5	0.041	Deterministic baseline target (variance factor 1.0)
enhanced_prediction	4-5	3-4	0	Enhanced ML with recent form, H2H, low-score analysis (confidence shrunk by projected_margin_is_one_run)
embedded_savant_run_environment	3-4	3-4	0	Target score adjusted by temporal Savant lineup quality versus opposing starter quality
fundamental_edge_target_review	3-4	3-4	0	Refreshed score target and win probability after Savant/pitcher/team-strength adjustments
capped_target	3-4	3-4	0	Capped at max 12 realistic runs
simulation_result	3-4	3-4	0	Final inning-by-inning simulation result

Largest player-level adjustments from the ultra-detailed trace

These are pregame model adjustments, not claims about actual postgame player error.

Team	Player	Role	Net effect	Largest change
DET	Kevin McGonigle	batter / 3B	worse	predicted_SO 2.396
WAS	CJ Abrams	batter / 2B	worse	predicted_RBI 2.327
DET	Javier Báez	batter / LF	better	predicted_H 2.189
DET	Drew Anderson	pitcher / Starter	better	predicted_SO 2
DET	Max Clark	batter / CF	better	predicted_H 1.98
DET	Ben Malgeri	batter / RF	worse	predicted_BB 1.663
DET	Spencer Torkelson	batter / 1B	worse	predicted_AB 1.651
DET	Eduardo Valencia	batter / C	worse	predicted_H -1.64

Structural audit findings

Severity	Rule	Finding	Subject
info	CURRENT_PITCHER_DUPLICATION	DET pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	DET
info	CURRENT_PITCHER_DUPLICATION	WAS pitcher list appears in both predicted_performance.pitchers and player_predictions_by_position.pitchers; IDs match and were de-duplicated in this audit.	WAS
info	CURRENT_FINALIZED_MATCH_BOX	DET model-prediction batting totals differ from final simulation box score (AB: current=31, sim=40; H: current=7, sim=10; 2B: current=1, sim=4; 3B: current=0, sim=1; SO: current=8, sim=14).	DET
info	CURRENT_PLAYER_MATCH_BOX	DET current finalized hitter lines are model predictions and differ from the simulation box score: Gleyber Torres: AB current=4 box=5, 2B current=0 box=1, RBI current=1 box=0, R current=1 box=0; Kevin McGonigle: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, SO current=0 box=3; Ben Malgeri: H current=1 box=0, BB current=0 box=2, SO current=1 box=2; Eduardo Valencia: AB current=4 box=5, H current=1 box=0, HR current=1 box=0, RBI current=1 box=0; Javier Báez: AB current=3 box=5, H current=1 box=3, 2B current=0 box=2, 3B current=0 box=1; Spencer Torkelson: AB current=3 box=5, BB current=1 box=0, SO current=1 box=0; Hao-Yu Lee: AB current=4 box=5, H current=1 box=2, 2B current=1 box=0, HR current=0 box=1; Max Clark: AB current=3 box=5, H current=1 box=3, 2B current=0 box=1, RBI current=0 box=1	DET
info	CURRENT_FINALIZED_MATCH_BOX	WAS model-prediction batting totals differ from final simulation box score (AB: current=31, sim=33; H: current=7, sim=3; RBI: current=4, sim=3; R: current=4, sim=3; BB: current=3, sim=7; SO: current=8, sim=9).	WAS
info	CURRENT_PLAYER_MATCH_BOX	WAS current finalized hitter lines are model predictions and differ from the simulation box score: Dylan Crews: H current=1 box=0, BB current=0 box=1, SO current=1 box=0; CJ Abrams: AB current=4 box=5, 2B current=1 box=0, HR current=0 box=1, RBI current=1 box=3; Yohandy Morales: AB current=3 box=4, H current=1 box=0, R current=1 box=0, SO current=1 box=0; Daylen Lile: H current=1 box=0, RBI current=1 box=0, R current=1 box=0, BB current=0 box=1; Nasim Nuñez: AB current=3 box=2, H current=0 box=1, R current=0 box=1, BB current=0 box=2; James Wood: H current=1 box=0, HR current=1 box=0, RBI current=1 box=0, R current=1 box=0; Andrés Chaparro: AB current=3 box=4, H current=1 box=0, RBI current=1 box=0, SO current=1 box=2; Keibert Ruiz: 2B current=0 box=1, R current=0 box=1, BB current=0 box=1, SO current=0 box=1	WAS
info	RECONCILIATION_PROPAGATION	No score_target_reconciliation step was used for this game.

Cross-game lever tension

Each candidate fix, and exactly which games it would help versus hurt. A lever that repairs one game and breaks two is visible here before it is proposed.

Day over day: did it improve or worsen

Compared against packet 20260921. Last recorded change: v4.30.0 (2026-09-08) Hitter shrinkage is derived from the between-hitter spread rather than fitted to slate error, so an established extreme profile is no longer flattened toward the league.

Metric	Prior	Current	Δ	Result
brier_score	0.2409	0.2448	+0.003936	regressed
accuracy	0.5833	0.563	-0.0203	regressed
team_run_mae	2.4917	2.4958	+0.0041	regressed
total_run_mae	3.2833	3.3109	+0.0276	regressed
calibration_error	0.0887	0.1034	+0.0147	regressed
starter_er_mae	1.6969	1.712	+0.0151	regressed
starter_outs_mae	3.5276	3.516	-0.0116	improved

Deep slate report — 20260922

Process, end to end

Slate summary

What drove the losses

Source coverage receipt

High-level causal issue register

Cross-slate trends (persistent-pattern memory)

Escalated persistent patterns

Finding streaks across the window

Metric direction across the window

Pipeline stage regression

Probability lineage and calibration

Score distribution and margin compression

Per-game evidence coverage

Every game, in depth (15)

ARI@COLHIT

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

CHW@KCHIT

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

CIN@ATLMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

CLE@BOSMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

HOU@SEAMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

LAA@OAKMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

MIA@CHCMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

MIL@PHIMISS

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

MIN@SFHIT

Player-level breakdown (projected vs actual)

Fundamental evidence entering the pick

Score and confidence evolution

Official postgame stage attribution

Largest player-level adjustments from the ultra-detailed trace

Structural audit findings

NYM@TEXMISS

Player-level breakdown (projected vs actual)